@elastic/schemas 0.4.4 → 0.5.0

package/src/es/json/inference._types.json

@@ -1 +1 @@
-{"$schema":"https://json-schema.org/draft/2020-12/schema","$defs":{"inference._types__EmbeddingContentType":{"enum":["text","image","audio","video","pdf"]},"inference._types__EmbeddingContentFormat":{"enum":["text","base64"]},"inference._types__CompletionInferenceResult":{"type":"object","properties":{"completion":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__CompletionResult"}}},"required":["completion"],"description":"Defines the completion result."},"inference._types__CompletionResult":{"type":"object","properties":{"result":{"type":"string"}},"required":["result"],"description":"The completion result object"},"inference._types__TaskType":{"enum":["sparse_embedding","text_embedding","rerank","completion","chat_completion","embedding"]},"inference._types__DeleteInferenceEndpointResult":{"type":"object","properties":{"acknowledged":{"type":"boolean","description":"For a successful response, this value is always true. On failure, an exception is returned instead."},"pipelines":{"type":"array","items":{"type":"string"}}},"required":["acknowledged","pipelines"],"description":"Acknowledged response. For dry_run, contains the list of pipelines which reference the inference endpoint"},"inference._types__EmbeddingInferenceResult":{"type":"object","properties":{"embeddings_bytes":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"embeddings_bits":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"embeddings":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingResult"}}},"description":"EmbeddingInferenceResult is an aggregation of mutually exclusive embeddings variants"},"inference._types__DenseEmbeddingByteResult":{"type":"object","properties":{"embedding":{"$ref":"./inference._types.json#/$defs/inference._types__DenseByteVector"}},"required":["embedding"],"description":"The dense embedding result object for byte representation"},"inference._types__DenseByteVector":{"type":"array","items":{"type":"number"},"description":"Dense Embedding results containing bytes are represented as Dense\nVectors of bytes."},"inference._types__DenseEmbeddingResult":{"type":"object","properties":{"embedding":{"$ref":"./inference._types.json#/$defs/inference._types__DenseVector"}},"required":["embedding"],"description":"The dense embedding result object for float representation"},"inference._types__DenseVector":{"type":"array","items":{"type":"number"},"description":"Dense Embedding results are represented as Dense Vectors\nof floats."},"inference._types__InferenceEndpointInfo":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskType","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"],"description":"Represents an inference endpoint as returned by the GET API"},"inference._types__InferenceChunkingSettings":{"type":"object","properties":{"max_chunk_size":{"type":"number","description":"The maximum size of a chunk in words.\nThis value cannot be lower than `20` (for `sentence` strategy) or `10` (for `word` strategy).\nThis value should not exceed the window size for the associated model."},"overlap":{"type":"number","description":"The number of overlapping words for chunks.\nIt is applicable only to a `word` chunking strategy.\nThis value cannot be higher than half the `max_chunk_size` value."},"sentence_overlap":{"type":"number","description":"The number of overlapping sentences for chunks.\nIt is applicable only for a `sentence` chunking strategy.\nIt can be either `1` or `0`."},"separator_group":{"type":"string","description":"Only applicable to the `recursive` strategy and required when using it.\n\nSets a predefined list of separators in the saved chunking settings based on the selected text type.\nValues can be `markdown` or `plaintext`.\n\nUsing this parameter is an alternative to manually specifying a custom `separators` list."},"separators":{"type":"array","items":{"type":"string"},"description":"Only applicable to the `recursive` strategy and required when using it.\n\nA list of strings used as possible split points when chunking text.\n\nEach string can be a plain string or a regular expression (regex) pattern.\nThe system tries each separator in order to split the text, starting from the first item in the list.\n\nAfter splitting, it attempts to recombine smaller pieces into larger chunks that stay within\nthe `max_chunk_size` limit, to reduce the total number of chunks generated."},"strategy":{"type":"string","description":"The chunking strategy: `sentence`, `word`, `none` or `recursive`.\n\n * If `strategy` is set to `recursive`, you must also specify:\n\n- `max_chunk_size`\n- either `separators` or`separator_group`\n\nLearn more about different chunking strategies in the linked documentation."}},"description":"Chunking configuration object"},"inference._types__RegionPolicyDoc":{"type":"object","properties":{"region_policy":{"$ref":"./inference._types.json#/$defs/inference._types__RegionPolicy"},"created_at":{"$ref":"./_types.json#/$defs/_types__DateTime","description":"The date and time the region policy was created."},"created_by":{"type":"string","description":"The user who created the region policy."},"updated_at":{"$ref":"./_types.json#/$defs/_types__DateTime","description":"The date and time the region policy was last updated."},"updated_by":{"type":"string","description":"The user who last updated the region policy."}},"required":["region_policy","created_at"],"description":"The stored region policy document."},"inference._types__RegionPolicy":{"type":"object","properties":{"allowed_geos":{"type":"array","items":{"type":"string"},"description":"The list of allowed geographic areas.\nMutually exclusive with `allowed_regions`."},"allowed_regions":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__CspRegion"},"description":"The list of allowed cloud service provider regions.\nMutually exclusive with `allowed_geos`."}},"description":"The region policy configuration.\nSpecify exactly one of `allowed_geos` or `allowed_regions`."},"inference._types__CspRegion":{"type":"object","properties":{"csp":{"type":"string","description":"The cloud service provider, for example `aws`, `gcp`, or `azure`."},"region":{"type":"string","description":"The region of the cloud service provider, for example `us-east-1`."}},"required":["csp","region"],"description":"A cloud service provider region."},"inference._types__InferenceResult":{"type":"object","properties":{"embeddings_bytes":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"embeddings_bits":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"embeddings":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingResult"}},"text_embedding_bytes":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"text_embedding_bits":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"text_embedding":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingResult"}},"sparse_embedding":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__SparseEmbeddingResult"}},"completion":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__CompletionResult"}},"rerank":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__RankedDocument"}}},"description":"InferenceResult is an aggregation of mutually exclusive variants"},"inference._types__SparseEmbeddingResult":{"type":"object","properties":{"is_truncated":{"type":"boolean","description":"Indicates if the text input was truncated in the request sent to the service"},"embedding":{"$ref":"./inference._types.json#/$defs/inference._types__SparseVector"}},"required":["is_truncated","embedding"]},"inference._types__SparseVector":{"type":"object","additionalProperties":{"type":"number"},"description":"Sparse Embedding tokens are represented as a dictionary\nof string to double."},"inference._types__RankedDocument":{"type":"object","properties":{"index":{"type":"number"},"relevance_score":{"type":"number"},"text":{"type":"string"}},"required":["index","relevance_score"],"description":"The rerank result object representing a single ranked document\nid: the original index of the document in the request\nrelevance_score: the relevance_score of the document relative to the query\ntext: Optional, the text of the document, if requested"},"inference._types__Ai21TaskType":{"enum":["completion","chat_completion"]},"inference._types__Ai21ServiceType":{"enum":["ai21"]},"inference._types__Ai21ServiceSettings":{"type":"object","properties":{"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the AI21 models documentation for the list of supported models and versions.\nService has been tested and confirmed to be working for `completion` and `chat_completion` tasks with the following models:\n* `jamba-mini`\n* `jamba-large`"},"api_key":{"type":"string","description":"A valid API key for accessing AI21 API.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the AI21 API.\nBy default, the `ai21` service sets the number of requests allowed per minute to 200. Please refer to AI21 documentation for more details."}},"required":["model_id"]},"inference._types__RateLimitSetting":{"type":"object","properties":{"requests_per_minute":{"type":"number","description":"The number of requests allowed per minute.\nBy default, the number of requests allowed per minute is set by each service as follows:\n\n* `alibabacloud-ai-search` service: `1000`\n* `amazonbedrock` service: `240`\n* `anthropic` service: `50`\n* `azureaistudio` service: `240`\n* `azureopenai` service and task type `text_embedding`: `1440`\n* `azureopenai` service and task types `completion` or `chat_completion`: `120`\n* `cohere` service: `10000`\n* `contextualai` service: `1000`\n* `elastic` service and task type `chat_completion`: `240`\n* `fireworksai` service: `6000`\n* `googleaistudio` service: `360`\n* `googlevertexai` service: `30000`\n* `hugging_face` service: `3000`\n* `jinaai` service: `2000`\n* `llama` service: `3000`\n* `mistral` service: `240`\n* `openai` service and task type `text_embedding`: `3000`\n* `openai` service and task type `completion`: `500`\n* `openshift_ai` service: `3000`\n* `voyageai` service: `2000`\n* `watsonxai` service: `120`"}},"description":"This setting helps to minimize the number of rate limit errors returned from the service."},"inference._types__InferenceEndpointInfoAi21":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAi21","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAi21":{"enum":["completion","chat_completion"]},"inference._types__AlibabaCloudTaskType":{"enum":["completion","rerank","sparse_embedding","text_embedding"]},"inference._types__AlibabaCloudServiceType":{"enum":["alibabacloud-ai-search"]},"inference._types__AlibabaCloudServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for the AlibabaCloud AI Search API."},"host":{"type":"string","description":"The name of the host address used for the inference task.\nYou can find the host address in the API keys section of the documentation."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from AlibabaCloud AI Search.\nBy default, the `alibabacloud-ai-search` service sets the number of requests allowed per minute to `1000`."},"service_id":{"type":"string","description":"The name of the model service to use for the inference task.\nThe following service IDs are available for the `completion` task:\n\n* `ops-qwen-turbo`\n* `qwen-turbo`\n* `qwen-plus`\n* `qwen-max ÷ qwen-max-longcontext`\n\nThe following service ID is available for the `rerank` task:\n\n* `ops-bge-reranker-larger`\n\nThe following service ID is available for the `sparse_embedding` task:\n\n* `ops-text-sparse-embedding-001`\n\nThe following service IDs are available for the `text_embedding` task:\n\n`ops-text-embedding-001`\n`ops-text-embedding-zh-001`\n`ops-text-embedding-en-001`\n`ops-text-embedding-002`"},"workspace":{"type":"string","description":"The name of the workspace used for the inference task."}},"required":["api_key","host","service_id","workspace"]},"inference._types__AlibabaCloudTaskSettings":{"type":"object","properties":{"input_type":{"type":"string","description":"For a `sparse_embedding` or `text_embedding` task, specify the type of input passed to the model.\nValid values are:\n\n* `ingest` for storing document embeddings in a vector database.\n* `search` for storing embeddings of search queries run against a vector database to find relevant documents."},"return_token":{"type":"boolean","description":"For a `sparse_embedding` task, it affects whether the token name will be returned in the response.\nIt defaults to `false`, which means only the token ID will be returned in the response."}}},"inference._types__InferenceEndpointInfoAlibabaCloudAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAlibabaCloudAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAlibabaCloudAI":{"enum":["text_embedding","rerank","completion","sparse_embedding"]},"inference._types__AmazonBedrockTaskType":{"enum":["chat_completion","completion","text_embedding"]},"inference._types__AmazonBedrockServiceType":{"enum":["amazonbedrock"]},"inference._types__AmazonBedrockServiceSettings":{"type":"object","properties":{"access_key":{"type":"string","description":"A valid AWS access key that has permissions to use Amazon Bedrock and access to models for inference requests."},"model":{"type":"string","description":"The base model ID or an ARN to a custom model based on a foundational model.\nThe base model IDs can be found in the Amazon Bedrock documentation.\nNote that the model ID must be available for the provider chosen and your IAM user must have access to the model."},"provider":{"type":"string","description":"The model provider for your deployment.\nNote that some providers may support only certain task types.\nSupported providers include:\n\n* `amazontitan` - available for `text_embedding` and `completion` task types\n* `anthropic` - available for `chat_completion` and `completion` task types\n* `ai21labs` - available for `chat_completion` and `completion` task types\n* `cohere` - available for `chat_completion`, `completion` and `text_embedding` task types\n* `meta` - available for `chat_completion` and `completion` task types\n* `mistral` - available for `chat_completion` and `completion` task types"},"region":{"type":"string","description":"The region that your model or ARN is deployed in.\nThe list of available regions per model can be found in the Amazon Bedrock documentation."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Amazon Bedrock.\nBy default, the `amazonbedrock` service sets the number of requests allowed per minute to 240."},"secret_key":{"type":"string","description":"A valid AWS secret key that is paired with the `access_key`.\nFor informationg about creating and managing access and secret keys, refer to the AWS documentation."}},"required":["access_key","model","region","secret_key"]},"inference._types__AmazonBedrockTaskSettings":{"type":"object","properties":{"max_new_tokens":{"type":"number","description":"For `chat_completion` and `completion` tasks, it sets the maximum number for the output tokens to be generated."},"temperature":{"type":"number","description":"For `chat_completion` and `completion` tasks, it is a number between 0.0 and 1.0 that controls the apparent creativity of the results.\nAt temperature 0.0 the model is most deterministic, at temperature 1.0 most random.\nIt should not be used if `top_p` or `top_k` is specified."},"top_k":{"type":"number","description":"For `chat_completion` and `completion` tasks, it limits samples to the top-K most likely words, balancing coherence and variability.\nIt is only available for anthropic, cohere, and mistral providers.\nIt is an alternative to `temperature`; it should not be used if `temperature` is specified."},"top_p":{"type":"number","description":"For `chat_completion` and `completion` tasks, it is a number in the range of 0.0 to 1.0, to eliminate low-probability tokens.\nTop-p uses nucleus sampling to select top tokens whose sum of likelihoods does not exceed a certain value, ensuring both variety and coherence.\nIt is an alternative to `temperature`; it should not be used if `temperature` is specified."}}},"inference._types__InferenceEndpointInfoAmazonBedrock":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAmazonBedrock","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAmazonBedrock":{"enum":["chat_completion","completion","text_embedding"]},"inference._types__TaskTypeAmazonSageMaker":{"enum":["text_embedding","completion","chat_completion","sparse_embedding","rerank"]},"inference._types__AmazonSageMakerServiceType":{"enum":["amazon_sagemaker"]},"inference._types__AmazonSageMakerServiceSettings":{"type":"object","properties":{"access_key":{"type":"string","description":"A valid AWS access key that has permissions to use Amazon SageMaker and access to models for invoking requests."},"endpoint_name":{"type":"string","description":"The name of the SageMaker endpoint."},"api":{"$ref":"./inference._types.json#/$defs/inference._types__AmazonSageMakerApi","description":"The API format to use when calling SageMaker.\nElasticsearch will convert the POST _inference request to this data format when invoking the SageMaker endpoint."},"region":{"type":"string","description":"The region that your endpoint or Amazon Resource Name (ARN) is deployed in.\nThe list of available regions per model can be found in the Amazon SageMaker documentation."},"secret_key":{"type":"string","description":"A valid AWS secret key that is paired with the `access_key`.\nFor information about creating and managing access and secret keys, refer to the AWS documentation."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__AmazonSageMakerSimilarity","description":"Required when `api` is `elastic` and task type is `text_embedding`. The similarity measure used when invoking the\n`text_embedding` task type."},"element_type":{"$ref":"./inference._types.json#/$defs/inference._types__AmazonSageMakerElementType","description":"Required when `api` is `elastic` and task type is `text_embedding`. The data type returned by the text embedding\nmodel.\nThis value is used when parsing the response back to Elasticsearch data structures."},"target_model":{"type":"string","description":"The model ID when calling a multi-model endpoint."},"target_container_hostname":{"type":"string","description":"The container to directly invoke when calling a multi-container endpoint."},"inference_component_name":{"type":"string","description":"The inference component to directly invoke when calling a multi-component endpoint."},"batch_size":{"type":"number","description":"The maximum number of inputs in each batch. This value is used by inference ingestion pipelines\nwhen processing semantic values. It correlates to the number of times the SageMaker endpoint is\ninvoked (one per batch of input)."},"dimensions":{"type":"number","description":"The number of dimensions returned by the text embedding models. If this value is not provided, then\nit is guessed by making invoking the endpoint for the `text_embedding` task."}},"required":["access_key","endpoint_name","api","region","secret_key"]},"inference._types__AmazonSageMakerApi":{"enum":["openai","elastic"]},"inference._types__AmazonSageMakerSimilarity":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__AmazonSageMakerElementType":{"enum":["byte","float","bit"]},"inference._types__AmazonSageMakerTaskSettings":{"type":"object","properties":{"custom_attributes":{"type":"string","description":"The AWS custom attributes passed verbatim through to the model running in the SageMaker Endpoint.\nValues will be returned in the `X-elastic-sagemaker-custom-attributes` header."},"enable_explanations":{"type":"string","description":"The optional JMESPath expression used to override the EnableExplanations provided during endpoint creation."},"inference_id":{"type":"string","description":"The capture data ID when enabled in the endpoint."},"session_id":{"type":"string","description":"The stateful session identifier for a new or existing session.\nNew sessions will be returned in the `X-elastic-sagemaker-new-session-id` header.\nClosed sessions will be returned in the `X-elastic-sagemaker-closed-session-id` header."},"target_variant":{"type":"string","description":"Specifies the variant when running with multi-variant Endpoints."}}},"inference._types__InferenceEndpointInfoAmazonSageMaker":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAmazonSageMaker","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__AnthropicTaskType":{"enum":["completion","chat_completion"]},"inference._types__AnthropicServiceType":{"enum":["anthropic"]},"inference._types__AnthropicServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for the Anthropic API."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Anthropic documentation for the list of supported models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Anthropic.\nBy default, the `anthropic` service sets the number of requests allowed per minute to 50."}},"required":["api_key","model_id"]},"inference._types__AnthropicTaskSettings":{"type":"object","properties":{"max_tokens":{"type":"number","description":"For a `completion` or `chat_completion` task, it is the maximum number of tokens to generate before stopping."},"temperature":{"type":"number","description":"For a `completion` or `chat_completion` task, it is the amount of randomness injected into the response.\nFor more details about the supported range, refer to Anthropic documentation."},"top_k":{"type":"number","description":"For a `completion` or `chat_completion` task, it specifies to only sample from the top K options for each subsequent token.\nIt is recommended for advanced use cases only.\nYou usually only need to use `temperature`."},"top_p":{"type":"number","description":"For a `completion` or `chat_completion` task, it specifies to use Anthropic's nucleus sampling.\nIn nucleus sampling, Anthropic computes the cumulative distribution over all the options for each subsequent token in decreasing probability order and cuts it off once it reaches the specified probability.\nYou should either alter `temperature` or `top_p`, but not both.\nIt is recommended for advanced use cases only.\nYou usually only need to use `temperature`."}},"required":["max_tokens"]},"inference._types__InferenceEndpointInfoAnthropic":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAnthropic","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAnthropic":{"enum":["completion","chat_completion"]},"inference._types__AzureAiStudioTaskType":{"enum":["completion","rerank","text_embedding"]},"inference._types__AzureAiStudioServiceType":{"enum":["azureaistudio"]},"inference._types__AzureAiStudioServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your Azure AI Studio model deployment.\nThis key can be found on the overview page for your deployment in the management section of your Azure AI Studio account.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"endpoint_type":{"type":"string","description":"The type of endpoint that is available for deployment through Azure AI Studio: `token` or `realtime`.\nThe `token` endpoint type is for \"pay as you go\" endpoints that are billed per token.\nThe `realtime` endpoint type is for \"real-time\" endpoints that are billed per hour of usage."},"target":{"type":"string","description":"The target URL of your Azure AI Studio model deployment.\nThis can be found on the overview page for your deployment in the management section of your Azure AI Studio account."},"provider":{"type":"string","description":"The model provider for your deployment.\nNote that some providers may support only certain task types.\nSupported providers include:\n\n* `cohere` - available for `text_embedding`, `rerank` and `completion` task types\n* `databricks` - available for `completion` task type only\n* `meta` - available for `completion` task type only\n* `microsoft_phi` - available for `completion` task type only\n* `mistral` - available for `completion` task type only\n* `openai` - available for `text_embedding` and `completion` task types"},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Azure AI Studio.\nBy default, the `azureaistudio` service sets the number of requests allowed per minute to 240."}},"required":["api_key","endpoint_type","target","provider"]},"inference._types__AzureAiStudioTaskSettings":{"type":"object","properties":{"do_sample":{"type":"number","description":"For a `completion` task, instruct the inference process to perform sampling.\nIt has no effect unless `temperature` or `top_p` is specified."},"max_new_tokens":{"type":"number","description":"For a `completion` task, provide a hint for the maximum number of output tokens to be generated."},"temperature":{"type":"number","description":"For a `completion` task, control the apparent creativity of generated completions with a sampling temperature.\nIt must be a number in the range of 0.0 to 2.0.\nIt should not be used if `top_p` is specified."},"top_p":{"type":"number","description":"For a `completion` task, make the model consider the results of the tokens with nucleus sampling probability.\nIt is an alternative value to `temperature` and must be a number in the range of 0.0 to 2.0.\nIt should not be used if `temperature` is specified."},"user":{"type":"string","description":"For a `text_embedding` task, specify the user issuing the request.\nThis information can be used for abuse detection."},"return_documents":{"type":"boolean","description":"For a `rerank` task, return doc text within the results."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents."}}},"inference._types__InferenceEndpointInfoAzureAIStudio":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAzureAIStudio","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAzureAIStudio":{"enum":["text_embedding","completion","rerank"]},"inference._types__AzureOpenAITaskType":{"enum":["completion","chat_completion","text_embedding"]},"inference._types__AzureOpenAIServiceType":{"enum":["azureopenai"]},"inference._types__AzureOpenAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Azure OpenAI account.\n\nIMPORTANT: You must specify either `api_key`, `entra_id`, or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint."},"api_version":{"type":"string","description":"The Azure API version ID to use.\nIt is recommended to use the latest supported non-preview version."},"client_id":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe application ID that's assigned to your app.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `tenant_id`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"client_secret":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe application secret that you created in the Microsoft app registration portal for your app.\n\nIMPORTANT: You must specify either `api_key`, `entra_id`, or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `tenant_id`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"deployment_id":{"type":"string","description":"The deployment name of your deployed models.\nYour Azure OpenAI deployments can be found though the Azure OpenAI Studio portal that is linked to your subscription."},"entra_id":{"type":"string","description":"A valid Microsoft Entra token.\n\nIMPORTANT: You must specify either `api_key`, `entra_id`, or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Azure.\nThe `azureopenai` service sets a default number of requests allowed per minute depending on the task type.\nFor `text_embedding`, it is set to `1440`.\nFor `completion` and `chat_completion`, it is set to `120`."},"resource_name":{"type":"string","description":"The name of your Azure OpenAI resource.\nYou can find this from the list of resources in the Azure Portal for your subscription."},"scopes":{"type":"array","items":{"type":"string"},"description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe resource identifier (application ID URI) of the resource you want, suffixed with .default\nFor example:\n```\n\"scopes\": [\n  \"https://cognitiveservices.azure.com/.default\"\n]\n```\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `tenant_id`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"tenant_id":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe directory tenant the application plans to operate against.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `tenant_id`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."}},"required":["api_version","deployment_id","resource_name"]},"inference._types__AzureOpenAITaskSettings":{"type":"object","properties":{"user":{"type":"string","description":"Specifies the user issuing the request.\nThis information can be used for abuse detection."},"headers":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies custom HTTP header parameters.\nFor example:\n```\n\"headers\": {\n  \"Custom-Header\": \"Some-Value\",\n  \"Another-Custom-Header\": \"Another-Value\"\n}\n```"}}},"inference._types__InferenceEndpointInfoAzureOpenAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAzureOpenAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAzureOpenAI":{"enum":["text_embedding","completion","chat_completion"]},"inference._types__CohereTaskType":{"enum":["completion","rerank","text_embedding"]},"inference._types__CohereServiceType":{"enum":["cohere"]},"inference._types__CohereServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Cohere account.\nYou can find or create your Cohere API keys on the Cohere API key settings page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"embedding_type":{"$ref":"./inference._types.json#/$defs/inference._types__CohereEmbeddingType","description":"For a `text_embedding` task, the types of embeddings you want to get back.\nUse `binary` for binary embeddings, which are encoded as bytes with signed int8 precision.\nUse `bit` for binary embeddings, which are encoded as bytes with signed int8 precision (this is a synonym of `binary`).\nUse `byte` for signed int8 embeddings (this is a synonym of `int8`).\nUse `float` for the default float embeddings.\nUse `int8` for signed int8 embeddings."},"model_id":{"type":"string","description":"For a `completion`, `rerank`, or `text_embedding` task, the name of the model to use for the inference task.\n\n* For the available `completion` models, refer to the [Cohere command docs](https://docs.cohere.com/docs/models#command).\n* For the available `rerank` models, refer to the [Cohere rerank docs](https://docs.cohere.com/reference/rerank-1).\n* For the available `text_embedding` models, refer to [Cohere embed docs](https://docs.cohere.com/reference/embed)."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Cohere.\nBy default, the `cohere` service sets the number of requests allowed per minute to 10000."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__CohereSimilarityType","description":"The similarity measure.\nIf the `embedding_type` is `float`, the default value is `dot_product`.\nIf the `embedding_type` is `int8` or `byte`, the default value is `cosine`."}},"required":["api_key","model_id"]},"inference._types__CohereEmbeddingType":{"enum":["binary","bit","byte","float","int8"]},"inference._types__CohereSimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__CohereTaskSettings":{"type":"object","properties":{"input_type":{"$ref":"./inference._types.json#/$defs/inference._types__CohereInputType","description":"For a `text_embedding` task, the type of input passed to the model.\nValid values are:\n\n* `classification`: Use it for embeddings passed through a text classifier.\n* `clustering`: Use it for the embeddings run through a clustering algorithm.\n* `ingest`: Use it for storing document embeddings in a vector database.\n* `search`: Use it for storing embeddings of search queries run against a vector database to find relevant documents.\n\nIMPORTANT: The `input_type` field is required when using embedding models `v3` and higher."},"return_documents":{"type":"boolean","description":"For a `rerank` task, return doc text within the results."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents.\nIf this inference endpoint is used in a `text_similarity_reranker` retriever query and `top_n` is set, it must be greater than or equal to `rank_window_size` in the query."},"truncate":{"$ref":"./inference._types.json#/$defs/inference._types__CohereTruncateType","description":"For a `text_embedding` task, the method to handle inputs longer than the maximum token length.\nValid values are:\n\n* `END`: When the input exceeds the maximum input token length, the end of the input is discarded.\n* `NONE`: When the input exceeds the maximum input token length, an error is returned.\n* `START`: When the input exceeds the maximum input token length, the start of the input is discarded."}},"required":["input_type"]},"inference._types__CohereInputType":{"enum":["classification","clustering","ingest","search"]},"inference._types__CohereTruncateType":{"enum":["END","NONE","START"]},"inference._types__InferenceEndpointInfoCohere":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeCohere","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeCohere":{"enum":["text_embedding","rerank","completion"]},"inference._types__TaskTypeContextualAI":{"enum":["rerank"]},"inference._types__ContextualAIServiceType":{"enum":["contextualai"]},"inference._types__ContextualAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Contexutual AI account.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Contextual AI documentation for the list of available rerank models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Contextual AI.\nThe `contextualai` service sets a default number of requests allowed per minute depending on the task type.\nFor `rerank`, it is set to `1000`."}},"required":["api_key","model_id"]},"inference._types__ContextualAITaskSettings":{"type":"object","properties":{"instruction":{"type":"string","description":"Instructions for the reranking model. Refer to <https://docs.contextual.ai/api-reference/rerank/rerank#body-instruction>\nOnly for the `rerank` task type."},"top_k":{"type":"number","description":"The number of most relevant documents to return.\nIf not specified, the reranking results of all documents will be returned.\nOnly for the `rerank` task type."}}},"inference._types__InferenceEndpointInfoContextualAi":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeContextualAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__CustomTaskType":{"enum":["text_embedding","sparse_embedding","rerank","completion"]},"inference._types__CustomServiceType":{"enum":["custom"]},"inference._types__CustomServiceSettings":{"type":"object","properties":{"batch_size":{"type":"number","description":"Specifies the batch size used for the semantic_text field. If the field is not provided, the default is 10.\nThe batch size is the maximum number of inputs in a single request to the upstream service.\nThe chunk within the batch are controlled by the selected chunking strategy for the semantic_text field."},"headers":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies the HTTP header parameters – such as `Authorization` or `Content-Type` – that are required to access the custom service.\nFor example:\n```\n\"headers\": {\n  \"Authorization\": \"Bearer ${api_key}\",\n  \"Content-Type\": \"application/json;charset=utf-8\"\n}\n```"},"input_type":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies the input type translation values that are used to replace the `${input_type}` template in the request body.\nFor example:\n```\n\"input_type\": {\n  \"translation\": {\n    \"ingest\": \"do_ingest\",\n    \"search\": \"do_search\"\n  },\n  \"default\": \"a_default\"\n},\n```\nIf the subsequent inference requests come from a search context, the `search` key will be used and the template will be replaced with `do_search`.\nIf it comes from the ingest context `do_ingest` is used. If it's a different context that is not specified, the default value will be used. If no default is specified an empty string is used.\n`translation` can be:\n* `classification`\n* `clustering`\n* `ingest`\n* `search`"},"query_parameters":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__CustomServiceQueryParameter"},"description":"Specifies the query parameters as a list of tuples. The arrays inside the `query_parameters` must have two items, a key and a value.\nFor example:\n```\n\"query_parameters\":[\n  [\"param_key\", \"some_value\"],\n  [\"param_key\", \"another_value\"],\n  [\"other_key\", \"other_value\"]\n]\n```\nIf the base url is `https://www.elastic.co` it results in: `https://www.elastic.co?param_key=some_value&param_key=another_value&other_key=other_value`."},"request":{"$ref":"./inference._types.json#/$defs/inference._types__CustomRequestParams","description":"The request configuration object."},"response":{"$ref":"./inference._types.json#/$defs/inference._types__CustomResponseParams","description":"The response configuration object."},"secret_parameters":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies secret parameters, like `api_key` or `api_token`, that are required to access the custom service.\nFor example:\n```\n\"secret_parameters\":{\n  \"api_key\":\"<api_key>\"\n}\n```"},"url":{"type":"string","description":"The URL endpoint to use for the requests."}},"required":["request","response","secret_parameters"]},"inference._types__CustomServiceQueryParameter":{"type":"array","items":{"type":"string"}},"inference._types__CustomRequestParams":{"type":"object","properties":{"content":{"type":"string","description":"The body structure of the request. It requires passing in the string-escaped result of the JSON format HTTP request body.\nFor example:\n```\n\"request\": \"{\\\"input\\\":${input}}\"\n```\n> info\n> The content string needs to be a single line except when using the Kibana console."}},"required":["content"]},"inference._types__CustomResponseParams":{"type":"object","properties":{"json_parser":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies the JSON parser that is used to parse the response from the custom service.\nDifferent task types require different json_parser parameters.\nFor example:\n```\n# text_embedding\n# For a response like this:\n\n{\n \"object\": \"list\",\n \"data\": [\n     {\n       \"object\": \"embedding\",\n       \"index\": 0,\n       \"embedding\": [\n           0.014539449,\n           -0.015288644\n       ]\n     }\n ],\n \"model\": \"text-embedding-ada-002-v2\",\n \"usage\": {\n     \"prompt_tokens\": 8,\n     \"total_tokens\": 8\n }\n}\n\n# the json_parser definition should look like this:\n\n\"response\":{\n  \"json_parser\":{\n    \"text_embeddings\":\"$.data[*].embedding[*]\"\n  }\n}\n\n# Elasticsearch supports the following embedding types:\n* float\n* byte\n* bit (or binary)\n\nTo specify the embedding type for the response, the `embedding_type`\nfield should be added in the `json_parser` object. Here's an example:\n\"response\":{\n  \"json_parser\":{\n    \"text_embeddings\":\"$.data[*].embedding[*]\",\n    \"embedding_type\":\"bit\"\n  }\n}\n\nIf `embedding_type` is not specified, it defaults to `float`.\n\n# sparse_embedding\n# For a response like this:\n\n{\n  \"request_id\": \"75C50B5B-E79E-4930-****-F48DBB392231\",\n  \"latency\": 22,\n  \"usage\": {\n     \"token_count\": 11\n  },\n  \"result\": {\n     \"sparse_embeddings\": [\n        {\n          \"index\": 0,\n          \"embedding\": [\n            {\n              \"token_id\": 6,\n              \"weight\": 0.101\n            },\n            {\n              \"token_id\": 163040,\n              \"weight\": 0.28417\n            }\n          ]\n        }\n     ]\n  }\n}\n\n# the json_parser definition should look like this:\n\n\"response\":{\n  \"json_parser\":{\n    \"token_path\":\"$.result.sparse_embeddings[*].embedding[*].token_id\",\n    \"weight_path\":\"$.result.sparse_embeddings[*].embedding[*].weight\"\n  }\n}\n\n# rerank\n# For a response like this:\n\n{\n  \"results\": [\n    {\n      \"index\": 3,\n      \"relevance_score\": 0.999071,\n      \"document\": \"abc\"\n    },\n    {\n      \"index\": 4,\n      \"relevance_score\": 0.7867867,\n      \"document\": \"123\"\n    },\n    {\n      \"index\": 0,\n      \"relevance_score\": 0.32713068,\n      \"document\": \"super\"\n    }\n  ],\n}\n\n# the json_parser definition should look like this:\n\n\"response\":{\n  \"json_parser\":{\n    \"reranked_index\":\"$.result.scores[*].index\",    // optional\n    \"relevance_score\":\"$.result.scores[*].score\",\n    \"document_text\":\"xxx\"    // optional\n  }\n}\n\n# completion\n# For a response like this:\n\n{\n \"id\": \"chatcmpl-B9MBs8CjcvOU2jLn4n570S5qMJKcT\",\n \"object\": \"chat.completion\",\n \"created\": 1741569952,\n \"model\": \"gpt-4.1-2025-04-14\",\n \"choices\": [\n   {\n    \"index\": 0,\n    \"message\": {\n      \"role\": \"assistant\",\n      \"content\": \"Hello! How can I assist you today?\",\n      \"refusal\": null,\n      \"annotations\": []\n    },\n    \"logprobs\": null,\n    \"finish_reason\": \"stop\"\n  }\n ]\n}\n\n# the json_parser definition should look like this:\n\n\"response\":{\n  \"json_parser\":{\n    \"completion_result\":\"$.choices[*].message.content\"\n  }\n}"}},"required":["json_parser"]},"inference._types__CustomTaskSettings":{"type":"object","properties":{"parameters":{"type":"object","additionalProperties":{"$ref":"./inference._types.json#/$defs/inference._types__CustomTaskParameter"},"description":"Specifies parameters that are required to run the custom service. The parameters depend on the model your custom service uses.\nFor example:\n```\n\"task_settings\":{\n  \"parameters\":{\n    \"input_type\":\"query\",\n    \"return_token\":true\n  }\n}\n```\n\n> warn\n> The `task_settings.parameters` cannot contain the same keys as `secret_parameters`. If they do, an error will be returned. This applies to PUT requests and POST requests."}}},"inference._types__CustomTaskParameter":{"oneOf":[{"type":"string"},{"type":"number"},{"type":"number"},{"type":"number"},{"type":"boolean"}]},"inference._types__InferenceEndpointInfoCustom":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeCustom","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeCustom":{"enum":["text_embedding","sparse_embedding","rerank","completion"]},"inference._types__TaskTypeDeepSeek":{"enum":["completion","chat_completion"]},"inference._types__DeepSeekServiceType":{"enum":["deepseek"]},"inference._types__DeepSeekServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your DeepSeek account.\nYou can find or create your DeepSeek API keys on the DeepSeek API key page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"model_id":{"type":"string","description":"For a `completion` or `chat_completion` task, the name of the model to use for the inference task.\n\nFor the available `completion` and `chat_completion` models, refer to the [DeepSeek Models & Pricing docs](https://api-docs.deepseek.com/quick_start/pricing)."},"url":{"type":"string","description":"The URL endpoint to use for the requests. Defaults to `https://api.deepseek.com/chat/completions`."}},"required":["api_key","model_id"]},"inference._types__InferenceEndpointInfoDeepSeek":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeDeepSeek","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__ElasticsearchTaskType":{"enum":["rerank","sparse_embedding","text_embedding"]},"inference._types__ElasticsearchServiceType":{"enum":["elasticsearch"]},"inference._types__ElasticsearchServiceSettings":{"type":"object","properties":{"adaptive_allocations":{"$ref":"./inference._types.json#/$defs/inference._types__AdaptiveAllocations","description":"Adaptive allocations configuration details.\nIf `enabled` is true, the number of allocations of the model is set based on the current load the process gets.\nWhen the load is high, a new model allocation is automatically created, respecting the value of `max_number_of_allocations` if it's set.\nWhen the load is low, a model allocation is automatically removed, respecting the value of `min_number_of_allocations` if it's set.\nIf `enabled` is true, do not set the number of allocations manually."},"deployment_id":{"type":"string","description":"The deployment identifier for a trained model deployment.\nWhen `deployment_id` is used the `model_id` is optional."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nIt can be the ID of a built-in model (for example, `.multilingual-e5-small` for E5) or a text embedding model that was uploaded by using the Eland client."},"num_allocations":{"type":"number","description":"The total number of allocations that are assigned to the model across machine learning nodes.\nIncreasing this value generally increases the throughput.\nIf adaptive allocations are enabled, do not set this value because it's automatically set."},"num_threads":{"type":"number","description":"The number of threads used by each model allocation during inference.\nThis setting generally increases the speed per inference request.\nThe inference process is a compute-bound process; `threads_per_allocations` must not exceed the number of available allocated processors per node.\nThe value must be a power of 2.\nThe maximum value is 32."},"long_document_strategy":{"type":"string","description":"Available only for the `rerank` task type using the Elastic reranker model.\nControls the strategy used for processing long documents during inference.\n\nPossible values:\n- `truncate` (default): Processes only the beginning of each document.\n- `chunk`: Splits long documents into smaller parts (chunks) before inference.\n\nWhen `long_document_strategy` is set to `chunk`, Elasticsearch splits each document into smaller parts but still returns a single score per document.\nThat score reflects the highest relevance score among all chunks."},"max_chunks_per_doc":{"type":"number","description":"Only for the `rerank` task type.\nLimits the number of chunks per document that are sent for inference when chunking is enabled.\nIf not set, all chunks generated for the document are processed."}},"required":["model_id","num_threads"]},"inference._types__AdaptiveAllocations":{"type":"object","properties":{"enabled":{"type":"boolean","description":"Turn on `adaptive_allocations`."},"max_number_of_allocations":{"type":"number","description":"The maximum number of allocations to scale to.\nIf set, it must be greater than or equal to `min_number_of_allocations`."},"min_number_of_allocations":{"type":"number","description":"The minimum number of allocations to scale to.\nIf set, it must be greater than or equal to 0.\nIf not defined, the deployment scales to 0."}}},"inference._types__ElasticsearchTaskSettings":{"type":"object","properties":{"return_documents":{"type":"boolean","description":"For a `rerank` task, return the document instead of only the index."}}},"inference._types__InferenceEndpointInfoElasticsearch":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeElasticsearch","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeElasticsearch":{"enum":["sparse_embedding","text_embedding","rerank"]},"inference._types__ElserTaskType":{"enum":["sparse_embedding"]},"inference._types__ElserServiceType":{"enum":["elser"]},"inference._types__ElserServiceSettings":{"type":"object","properties":{"adaptive_allocations":{"$ref":"./inference._types.json#/$defs/inference._types__AdaptiveAllocations","description":"Adaptive allocations configuration details.\nIf `enabled` is true, the number of allocations of the model is set based on the current load the process gets.\nWhen the load is high, a new model allocation is automatically created, respecting the value of `max_number_of_allocations` if it's set.\nWhen the load is low, a model allocation is automatically removed, respecting the value of `min_number_of_allocations` if it's set.\nIf `enabled` is true, do not set the number of allocations manually."},"num_allocations":{"type":"number","description":"The total number of allocations this model is assigned across machine learning nodes.\nIncreasing this value generally increases the throughput.\nIf adaptive allocations is enabled, do not set this value because it's automatically set."},"num_threads":{"type":"number","description":"The number of threads used by each model allocation during inference.\nIncreasing this value generally increases the speed per inference request.\nThe inference process is a compute-bound process; `threads_per_allocations` must not exceed the number of available allocated processors per node.\nThe value must be a power of 2.\nThe maximum value is 32.\n\n> info\n> If you want to optimize your ELSER endpoint for ingest, set the number of threads to 1. If you want to optimize your ELSER endpoint for search, set the number of threads to greater than 1."}},"required":["num_allocations","num_threads"]},"inference._types__InferenceEndpointInfoELSER":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeELSER","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeELSER":{"enum":["sparse_embedding"]},"inference._types__FireworksAITaskType":{"enum":["chat_completion","completion","text_embedding"]},"inference._types__FireworksAIServiceType":{"enum":["fireworksai"]},"inference._types__FireworksAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Fireworks AI account.\nYou can find or create your API keys in the Fireworks AI dashboard.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Fireworks AI documentation for the list of available models for chat completion, completion, and text embedding.\nFor text embedding, supported models include the Qwen3 embedding family (e.g. `fireworks/qwen3-embedding-8b`) and other models in the Fireworks model library."},"url":{"type":"string","description":"The URL endpoint to use for the requests.\nIf not provided, the default Fireworks AI API endpoint is used."},"dimensions":{"type":"number","description":"For a `text_embedding` task, the number of dimensions the resulting output embeddings should have.\nVariable-length embeddings are supported via this parameter."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__FireworksAISimilarityType","description":"For a `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Fireworks AI API.\nRate limit grouping is per API key only.\nBy default, the `fireworksai` service sets the number of requests allowed per minute to 6000."}},"required":["api_key","model_id"]},"inference._types__FireworksAISimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__FireworksAITaskSettings":{"type":"object","properties":{"user":{"type":"string","description":"For a `completion` or`chat_completion` task, specify the user issuing the request.\nThis information can be used for abuse detection."},"headers":{"type":"object","additionalProperties":{"type":"string"},"description":"For a `completion` or`chat_completion` task. Specifies custom HTTP header parameters.\nFor example:\n```\n\"headers\": {\n  \"Custom-Header\": \"Some-Value\",\n  \"Another-Custom-Header\": \"Another-Value\"\n}\n```"}}},"inference._types__InferenceEndpointInfoFireworksAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeFireworksAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeFireworksAI":{"enum":["chat_completion","completion","text_embedding"]},"inference._types__GoogleAiStudioTaskType":{"enum":["completion","text_embedding"]},"inference._types__GoogleAiServiceType":{"enum":["googleaistudio"]},"inference._types__GoogleAiStudioServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your Google Gemini account."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Google documentation for the list of supported models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Google AI Studio.\nBy default, the `googleaistudio` service sets the number of requests allowed per minute to 360."}},"required":["api_key","model_id"]},"inference._types__InferenceEndpointInfoGoogleAIStudio":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeGoogleAIStudio","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeGoogleAIStudio":{"enum":["text_embedding","completion"]},"inference._types__GoogleVertexAITaskType":{"enum":["rerank","text_embedding","completion","chat_completion"]},"inference._types__GoogleVertexAIServiceType":{"enum":["googlevertexai"]},"inference._types__GoogleVertexAIServiceSettings":{"type":"object","properties":{"provider":{"$ref":"./inference._types.json#/$defs/inference._types__GoogleModelGardenProvider","description":"The name of the Google Model Garden Provider for `completion` and `chat_completion` tasks.\nIn order for a Google Model Garden endpoint to be used `provider` must be defined and be other than `google`.\nModes:\n- Google Model Garden (third-party models): set `provider` to a supported non-`google` value and provide `url` and/or `streaming_url`.\n- Google Vertex AI: omit `provider` or set it to `google`. In this mode, do not set `url` or `streaming_url` and Elastic will construct the endpoint url from `location`, `model_id`, and `project_id` parameters."},"url":{"type":"string","description":"The URL for non-streaming `completion` requests to a Google Model Garden provider endpoint.\nIf both `url` and `streaming_url` are provided, each is used for its respective mode.\nIf `streaming_url` is not provided, `url` is also used for streaming `completion` and `chat_completion`.\nIf `provider` is not provided or set to `google` (Google Vertex AI), do not set `url` (or `streaming_url`).\nAt least one of `url` or `streaming_url` must be provided for Google Model Garden endpoint usage.\nCertain providers require separate URLs for streaming and non-streaming operations (e.g., Anthropic, Mistral, AI21). Others support both operation types through a single URL (e.g., Meta, Hugging Face).\nInformation on constructing the URL for various providers can be found in the Google Model Garden documentation for the model, or on the endpoint’s `Sample request` page. The request examples also illustrate the proper formatting for the `url`."},"streaming_url":{"type":"string","description":"The URL for streaming `completion` and `chat_completion` requests to a Google Model Garden provider endpoint.\nIf both `streaming_url` and `url` are provided, each is used for its respective mode.\nIf `url` is not provided, `streaming_url` is also used for non-streaming `completion` requests.\nIf `provider` is not provided or set to `google` (Google Vertex AI), do not set `streaming_url` (or `url`).\nAt least one of `streaming_url` or `url` must be provided for Google Model Garden endpoint usage.\nCertain providers require separate URLs for streaming and non-streaming operations (e.g., Anthropic, Mistral, AI21). Others support both operation types through a single URL (e.g., Meta, Hugging Face).\nInformation on constructing the URL for various providers can be found in the Google Model Garden documentation for the model, or on the endpoint’s `Sample request` page. The request examples also illustrate the proper formatting for the `streaming_url`."},"location":{"type":"string","description":"The name of the location to use for the inference task for the Google Vertex AI inference task.\nFor Google Vertex AI, when `provider` is omitted or `google`, `location` is optional.\nWhen `location` is omitted, the request targets the Google Vertex AI global endpoint (`https://aiplatform.googleapis.com`); when it is provided, the corresponding regional endpoint (`https://<location>-aiplatform.googleapis.com`) is used.\nThe field must be omitted to select the global endpoint; an empty string is not a valid `location`.\nFor Google Model Garden's `completion` and `chat_completion` tasks, when `provider` is a supported non-`google` value - `location` is ignored.\nRefer to the Google documentation for the list of supported locations."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nFor Google Vertex AI `model_id` is mandatory.\nFor Google Model Garden's `completion` and `chat_completion` tasks, when `provider` is a supported non-`google` value - `model_id` will be used for some providers that require it, otherwise - ignored.\nRefer to the Google documentation for the list of supported models for Google Vertex AI."},"project_id":{"type":"string","description":"The name of the project to use for the Google Vertex AI inference task.\nFor Google Vertex AI `project_id` is mandatory.\nFor Google Model Garden's `completion` and `chat_completion` tasks, when `provider` is a supported non-`google` value - `project_id` is ignored."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Google Vertex AI.\nBy default, the `googlevertexai` service sets the number of requests allowed per minute to 30.000."},"service_account_json":{"type":"string","description":"A valid service account in JSON format for the Google Vertex AI API."},"dimensions":{"type":"number","description":"For a `text_embedding` task, the number of dimensions the resulting output embeddings should have.\nBy default, the model's standard output dimension is used.\nRefer to the Google documentation for more information."},"max_batch_size":{"type":"number","description":"Only applicable for the `text_embedding` task type.\nControls the batch size of chunked inference requests sent to Google Vertex AI.\n\nSetting this parameter lower reduces the risk of exceeding token limits but may result in more API calls. Setting it higher increases throughput but may risk hitting token limits.\n\nTo estimate a safe `max_batch_size` value, you can use it together with the `max_chunk_size` parameter using the following formula:\n`max_batch_size ≈ max_chunk_size × 1.3 × 512 ÷ 20000`\n\nWhere:\n- `1.3` is an approximate tokens-per-word ratio\n- `512` is the maximum number of chunks that can be generated per document\n- `20000` is the Google Vertex AI token limit per request\n\nThis estimate assumes the worst-case scenario with a document generating the maximum 512 chunks."}},"required":["service_account_json"]},"inference._types__GoogleModelGardenProvider":{"enum":["google","anthropic","meta","hugging_face","mistral","ai21"]},"inference._types__GoogleVertexAITaskSettings":{"type":"object","properties":{"auto_truncate":{"type":"boolean","description":"For a `text_embedding` task, truncate inputs longer than the maximum token length automatically."},"top_n":{"type":"number","description":"For a `rerank` task, the number of the top N documents that should be returned."},"thinking_config":{"$ref":"./inference._types.json#/$defs/inference._types__ThinkingConfig","description":"For a `completion` or `chat_completion` task, allows configuration of the thinking features for the model.\nRefer to the Google documentation for the allowable configurations for each model type."},"max_tokens":{"type":"number","description":"For `completion` and `chat_completion` tasks, specifies the `max_tokens` value for requests sent to the Google Model Garden `anthropic` provider.\nIf `provider` is not set to `anthropic`, this field is ignored.\nIf `max_tokens` is specified - it must be a positive integer. If not specified, the default value of 1024 is used.\nAnthropic models require `max_tokens` to be set for each request. Please refer to the Anthropic documentation for more information."}}},"inference._types__ThinkingConfig":{"type":"object","properties":{"thinking_budget":{"type":"number","description":"Indicates the desired thinking budget in tokens."}}},"inference._types__InferenceEndpointInfoGoogleVertexAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeGoogleVertexAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeGoogleVertexAI":{"enum":["chat_completion","completion","text_embedding","rerank"]},"inference._types__GroqTaskType":{"enum":["chat_completion"]},"inference._types__GroqServiceType":{"enum":["groq"]},"inference._types__GroqServiceSettings":{"type":"object","properties":{"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Groq model documentation for the list of supported models and versions.\nService has been tested and confirmed to be working for `completion` and `chat_completion` tasks with the following models:\n* `llama-3.3-70b-versatile`"},"api_key":{"type":"string","description":"A valid API key for accessing Groq API.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Groq API.\nBy default, the `groq` service sets the number of requests allowed per minute to 1000. Refer to Groq documentation for more details."}},"required":["model_id"]},"inference._types__InferenceEndpointInfoGroq":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeGroq","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeGroq":{"enum":["chat_completion"]},"inference._types__HuggingFaceTaskType":{"enum":["chat_completion","completion","rerank","text_embedding"]},"inference._types__HuggingFaceServiceType":{"enum":["hugging_face"]},"inference._types__HuggingFaceServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid access token for your HuggingFace account.\nYou can create or find your access tokens on the HuggingFace settings page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Hugging Face.\nBy default, the `hugging_face` service sets the number of requests allowed per minute to 3000 for all supported tasks.\nHugging Face does not publish a universal rate limit — actual limits may vary.\nIt is recommended to adjust this value based on the capacity and limits of your specific deployment environment."},"url":{"type":"string","description":"The URL endpoint to use for the requests.\nFor `completion` and `chat_completion` tasks, the deployed model must be compatible with the Hugging Face Chat Completion interface (see the linked external documentation for details). The endpoint URL for the request must include `/v1/chat/completions`.\nIf the model supports the OpenAI Chat Completion schema, a toggle should appear in the interface. Enabling this toggle doesn't change any model behavior, it reveals the full endpoint URL needed (which should include `/v1/chat/completions`) when configuring the inference endpoint in Elasticsearch. If the model doesn't support this schema, the toggle may not be shown."},"model_id":{"type":"string","description":"The name of the HuggingFace model to use for the inference task.\nFor `completion` and `chat_completion` tasks, this field is optional but may be required for certain models — particularly when using serverless inference endpoints.\nFor the `text_embedding` task, this field should not be included. Otherwise, the request will fail."}},"required":["api_key","url"]},"inference._types__HuggingFaceTaskSettings":{"type":"object","properties":{"return_documents":{"type":"boolean","description":"For a `rerank` task, return doc text within the results."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents."}}},"inference._types__InferenceEndpointInfoHuggingFace":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeHuggingFace","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeHuggingFace":{"enum":["chat_completion","completion","rerank","text_embedding"]},"inference._types__JinaAITaskType":{"enum":["embedding","rerank","text_embedding"]},"inference._types__JinaAIServiceType":{"enum":["jinaai"]},"inference._types__JinaAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your JinaAI account.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"model_id":{"type":"string","description":"The name of the model to use for the inference task."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from JinaAI.\nBy default, the `jinaai` service sets the number of requests allowed per minute to 2000 for all task types."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__JinaAISimilarityType","description":"For an `embedding` or `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm.\nThe default values varies with the embedding type.\nFor example, a float embedding type uses a `dot_product` similarity measure by default."},"dimensions":{"type":"number","description":"For an `embedding` or `text_embedding` task, the number of dimensions the resulting output embeddings should have.\nBy default, the model's standard output dimension is used.\nRefer to the Jina documentation for more information."},"embedding_type":{"$ref":"./inference._types.json#/$defs/inference._types__JinaAIElementType","description":"For an `embedding` or `text_embedding` task, the data type returned by the model.\nUse `bit` for binary embeddings, which are encoded as bytes with signed int8 precision.\nUse `binary` for binary embeddings, which are encoded as bytes with signed int8 precision (this is a synonym of `bit`).\nUse `float` for the default float embeddings."},"multimodal_model":{"type":"boolean","description":"For the `embedding` task, whether the model supports multimodal inputs. If true, requests sent to the Jina model\nwill use the multimodal request format (a list of objects). If false, requests sent to the model will use the same\nformat as the `text_embedding` task (a list of strings). Setting this to `false` allows the `embedding` task to be\nused with models that do not support multimodal requests."}},"required":["api_key","model_id"]},"inference._types__JinaAISimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__JinaAIElementType":{"enum":["binary","bit","float"]},"inference._types__JinaAITaskSettings":{"type":"object","properties":{"return_documents":{"type":"boolean","description":"For a `rerank` task, return the doc text within the results."},"input_type":{"$ref":"./inference._types.json#/$defs/inference._types__JinaAITextEmbeddingTask","description":"For an `embedding` or `text_embedding` task, the task passed to the model.\nValid values are:\n\n* `classification`: Use it for embeddings passed through a classifier.\n* `clustering`: Use it for the embeddings run through a clustering algorithm.\n* `ingest`: Use it for storing document embeddings in a vector database.\n* `search`: Use it for storing embeddings of search queries run against a vector database to find relevant documents."},"late_chunking":{"type":"boolean","description":"For an `embedding` or `text_embedding` task, controls when text is split into chunks.\nWhen set to `true`, a request from Elasticsearch contains only chunks related to a single document. Instead of batching chunks across documents, Elasticsearch sends them in separate requests. This ensures that chunk embeddings retain context from the entire document, improving semantic quality.\n\nIf a document exceeds the model's context limits, or if the document contains non-text inputs (relevant when using the multimodal `embedding` task), late chunking is automatically disabled for that document only and standard chunking is used instead.\n\nIf not specified, defaults to `false`."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents.\nIf this inference endpoint is used in a `text_similarity_reranker` retriever query and `top_n` is set, it must be greater than or equal to `rank_window_size` in the query."}}},"inference._types__JinaAITextEmbeddingTask":{"enum":["classification","clustering","ingest","search"]},"inference._types__InferenceEndpointInfoJinaAi":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeJinaAi","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeJinaAi":{"enum":["embedding","text_embedding","rerank"]},"inference._types__LlamaTaskType":{"enum":["text_embedding","completion","chat_completion"]},"inference._types__LlamaServiceType":{"enum":["llama"]},"inference._types__LlamaServiceSettings":{"type":"object","properties":{"url":{"type":"string","description":"The URL endpoint of the Llama stack endpoint.\nURL must contain:\n* For `text_embedding` task - `/v1/inference/embeddings`.\n* For `completion` and `chat_completion` tasks - `/v1/openai/v1/chat/completions`."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Llama downloading models documentation for different ways of getting a list of available models and downloading them.\nService has been tested and confirmed to be working with the following models:\n* For `text_embedding` task - `all-MiniLM-L6-v2`.\n* For `completion` and `chat_completion` tasks - `llama3.2:3b`."},"max_input_tokens":{"type":"number","description":"For a `text_embedding` task, the maximum number of tokens per input before chunking occurs."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__LlamaSimilarityType","description":"For a `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Llama API.\nBy default, the `llama` service sets the number of requests allowed per minute to 3000."}},"required":["url","model_id"]},"inference._types__LlamaSimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__InferenceEndpointInfoLlama":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeLlama","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeLlama":{"enum":["text_embedding","chat_completion","completion"]},"inference._types__MistralTaskType":{"enum":["text_embedding","completion","chat_completion"]},"inference._types__MistralServiceType":{"enum":["mistral"]},"inference._types__MistralServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your Mistral account.\nYou can find your Mistral API keys or you can create a new one on the API Keys page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"max_input_tokens":{"type":"number","description":"The maximum number of tokens per input before chunking occurs."},"model":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Mistral models documentation for the list of available models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Mistral API.\nBy default, the `mistral` service sets the number of requests allowed per minute to 240."}},"required":["api_key","model"]},"inference._types__InferenceEndpointInfoMistral":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeMistral","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeMistral":{"enum":["text_embedding","chat_completion","completion"]},"inference._types__NvidiaTaskType":{"enum":["chat_completion","completion","rerank","text_embedding"]},"inference._types__NvidiaServiceType":{"enum":["nvidia"]},"inference._types__NvidiaServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Nvidia endpoint.\nCan be found in `API Keys` section of Nvidia account settings."},"url":{"type":"string","description":"The URL of the Nvidia model endpoint. If not provided, the default endpoint URL is used depending on the task type:\n\n* For `text_embedding` task - `https://integrate.api.nvidia.com/v1/embeddings`.\n* For `completion` and `chat_completion` tasks - `https://integrate.api.nvidia.com/v1/chat/completions`.\n* For `rerank` task - `https://ai.api.nvidia.com/v1/retrieval/nvidia/reranking`."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the model's documentation for the name if needed.\nService has been tested and confirmed to be working with the following models:\n\n* For `text_embedding` task - `nvidia/llama-3.2-nv-embedqa-1b-v2`.\n* For `completion` and `chat_completion` tasks - `microsoft/phi-3-mini-128k-instruct`.\n* For `rerank` task - `nv-rerank-qa-mistral-4b:1`.\nService doesn't support `text_embedding` task `baai/bge-m3` and `nvidia/nvclip` models due to them not recognizing the `input_type` parameter."},"max_input_tokens":{"type":"number","description":"For a `text_embedding` task, the maximum number of tokens per input. Inputs exceeding this value are truncated prior to sending to the Nvidia API."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__NvidiaSimilarityType","description":"For a `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Nvidia API.\nBy default, the `nvidia` service sets the number of requests allowed per minute to 3000."}},"required":["api_key","model_id"]},"inference._types__NvidiaSimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__NvidiaTaskSettings":{"type":"object","properties":{"input_type":{"$ref":"./inference._types.json#/$defs/inference._types__NvidiaInputType","description":"For a `text_embedding` task, type of input sent to the Nvidia endpoint.\nValid values are:\n\n* `ingest`: Mapped to Nvidia's `passage` value in request. Used when generating embeddings during indexing.\n* `search`: Mapped to Nvidia's `query` value in request. Used when generating embeddings during querying.\n\nIMPORTANT: For Nvidia endpoints, if the `input_type` field is not specified, it defaults to `query`."},"truncate":{"$ref":"./inference._types.json#/$defs/inference._types__CohereTruncateType","description":"For a `text_embedding` task, the method used by the Nvidia model to handle inputs longer than the maximum token length.\nValid values are:\n\n* `END`: When the input exceeds the maximum input token length, the end of the input is discarded.\n* `NONE`: When the input exceeds the maximum input token length, an error is returned.\n* `START`: When the input exceeds the maximum input token length, the start of the input is discarded."}}},"inference._types__NvidiaInputType":{"enum":["ingest","search"]},"inference._types__InferenceEndpointInfoNvidia":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference ID"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeNvidia","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeNvidia":{"enum":["chat_completion","completion","rerank","text_embedding"]},"inference._types__OpenAITaskType":{"enum":["chat_completion","completion","text_embedding","embedding"]},"inference._types__OpenAIServiceType":{"enum":["openai"]},"inference._types__OpenAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your OpenAI account.\nYou can find your OpenAI API keys in your OpenAI account under the API keys section.\n\nIMPORTANT: You must specify either `api_key` or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint."},"client_id":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe application ID that's assigned to your app.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `token_url`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"client_secret":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe application secret that you created for your app.\n\nIMPORTANT: You must specify either `api_key` or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `token_url`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"dimensions":{"type":"number","description":"For a `text_embedding` or `embedding` task, the number of dimensions the resulting output embeddings should have.\nIt is supported only in `text-embedding-3` and later models.\nIf it is not set, the OpenAI defined default for the model is used."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the OpenAI documentation for the list of available text embedding models."},"organization_id":{"type":"string","description":"The unique identifier for your organization.\nYou can find the Organization ID in your OpenAI account under *Settings > Organizations*."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from OpenAI.\nThe `openai` service sets a default number of requests allowed per minute depending on the task type.\nFor `text_embedding` and `embedding`, it is set to `3000`.\nFor `completion` and `chat_completion`, it is set to `500`."},"scopes":{"type":"array","items":{"type":"string"},"description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe resource identifier of the resource you want.\nFor example:\n```\n\"scopes\": [\n  \"scope1\",\n  \"scope2\"\n]\n```\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `token_url`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__OpenAISimilarityType","description":"For a `text_embedding` or `embedding` task, the similarity measure. One of `cosine`, `dot_product`, `l2_norm`. Defaults to `dot_product`."},"token_url":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nAn OAuth2 token endpoint where Elasticsearch sends a request to exchange client credentials for an access token.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `token_url`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"url":{"type":"string","description":"The URL endpoint to use for the requests.\nIt can be changed for testing purposes.\nDefault value is `https://api.openai.com/v1/embeddings` for a `text_embedding` or `embedding` task,\n`https://api.openai.com/v1/chat/completions` for a `completion` or `chat_completion` task."}},"required":["model_id"]},"inference._types__OpenAISimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__OpenAITaskSettings":{"type":"object","properties":{"user":{"type":"string","description":"Specifies the user issuing the request.\nThis information can be used for abuse detection."},"headers":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies custom HTTP header parameters.\nFor example:\n```\n\"headers\": {\n  \"Custom-Header\": \"Some-Value\",\n  \"Another-Custom-Header\": \"Another-Value\"\n}\n```"}}},"inference._types__InferenceEndpointInfoOpenAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeOpenAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeOpenAI":{"enum":["text_embedding","chat_completion","completion","embedding"]},"inference._types__OpenShiftAiTaskType":{"enum":["text_embedding","completion","chat_completion","rerank"]},"inference._types__OpenShiftAiServiceType":{"enum":["openshift_ai"]},"inference._types__OpenShiftAiServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your OpenShift AI endpoint.\nCan be found in `Token authentication` section of model related information."},"url":{"type":"string","description":"The URL of the OpenShift AI hosted model endpoint."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the hosted model's documentation for the name if needed.\nService has been tested and confirmed to be working with the following models:\n* For `text_embedding` task - `gritlm-7b`.\n* For `completion` and `chat_completion` tasks - `llama-31-8b-instruct`.\n* For `rerank` task - `bge-reranker-v2-m3`."},"max_input_tokens":{"type":"number","description":"For a `text_embedding` task, the maximum number of tokens per input before chunking occurs."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__OpenShiftAiSimilarityType","description":"For a `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm.\nIf not specified, the default dot_product value is used."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the OpenShift AI API.\nBy default, the `openshift_ai` service sets the number of requests allowed per minute to 3000."}},"required":["api_key","url"]},"inference._types__OpenShiftAiSimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__OpenShiftAiTaskSettings":{"type":"object","properties":{"return_documents":{"type":"boolean","description":"For a `rerank` task, whether to return the source documents in the response."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return."}}},"inference._types__InferenceEndpointInfoOpenShiftAi":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeOpenShiftAi","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeOpenShiftAi":{"enum":["text_embedding","chat_completion","completion","rerank"]},"inference._types__VoyageAITaskType":{"enum":["text_embedding","rerank"]},"inference._types__VoyageAIServiceType":{"enum":["voyageai"]},"inference._types__VoyageAIServiceSettings":{"type":"object","properties":{"dimensions":{"type":"number","description":"The number of dimensions for resulting output embeddings.\nThis setting maps to `output_dimension` in the VoyageAI documentation.\nOnly for the `text_embedding` task type."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the VoyageAI documentation for the list of available text embedding and rerank models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from VoyageAI.\nThe `voyageai` service sets a default number of requests allowed per minute depending on the task type.\nFor both `text_embedding` and `rerank`, it is set to `2000`."},"embedding_type":{"type":"number","description":"The data type for the embeddings to be returned.\nThis setting maps to `output_dtype` in the VoyageAI documentation.\nPermitted values: float, int8, bit.\n`int8` is a synonym of `byte` in the VoyageAI documentation.\n`bit` is a synonym of `binary` in the VoyageAI documentation.\nOnly for the `text_embedding` task type."}},"required":["model_id"]},"inference._types__VoyageAITaskSettings":{"type":"object","properties":{"input_type":{"type":"string","description":"Type of the input text.\nPermitted values: `ingest` (maps to `document` in the VoyageAI documentation), `search` (maps to `query` in the VoyageAI documentation).\nOnly for the `text_embedding` task type."},"return_documents":{"type":"boolean","description":"Whether to return the source documents in the response.\nOnly for the `rerank` task type."},"top_k":{"type":"number","description":"The number of most relevant documents to return.\nIf not specified, the reranking results of all documents will be returned.\nOnly for the `rerank` task type."},"truncation":{"type":"boolean","description":"Whether to truncate the input texts to fit within the context length."}}},"inference._types__InferenceEndpointInfoVoyageAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeVoyageAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeVoyageAI":{"enum":["text_embedding","rerank"]},"inference._types__WatsonxTaskType":{"enum":["text_embedding","rerank","chat_completion","completion"]},"inference._types__WatsonxServiceType":{"enum":["watsonxai"]},"inference._types__WatsonxServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your Watsonx account.\nYou can find your Watsonx API keys or you can create a new one on the API keys page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"api_version":{"type":"string","description":"A version parameter that takes a version date in the format of `YYYY-MM-DD`.\nFor the active version data parameters, refer to the Wastonx documentation."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the IBM Embedding Models section in the Watsonx documentation for the list of available text embedding models.\nRefer to the IBM library - Foundation models in Watsonx.ai."},"project_id":{"type":"string","description":"The identifier of the IBM Cloud project to use for the inference task."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Watsonx.\nBy default, the `watsonxai` service sets the number of requests allowed per minute to 120."},"url":{"type":"string","description":"The URL of the inference endpoint that you created on Watsonx."}},"required":["api_key","api_version","model_id","project_id","url"]},"inference._types__InferenceEndpointInfoWatsonx":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeWatsonx","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeWatsonx":{"enum":["text_embedding","chat_completion","completion"]},"inference._types__RerankedInferenceResult":{"type":"object","properties":{"rerank":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__RankedDocument"}}},"required":["rerank"],"description":"Defines the response for a rerank request."},"inference._types__SparseEmbeddingInferenceResult":{"type":"object","properties":{"sparse_embedding":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__SparseEmbeddingResult"}}},"required":["sparse_embedding"],"description":"The response format for the sparse embedding request."},"inference._types__TextEmbeddingInferenceResult":{"type":"object","properties":{"text_embedding_bytes":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"text_embedding_bits":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"text_embedding":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingResult"}}},"description":"TextEmbeddingInferenceResult is an aggregation of mutually exclusive text_embedding variants"}}}
+{"$schema":"https://json-schema.org/draft/2020-12/schema","$defs":{"inference._types__EmbeddingContentType":{"enum":["text","image","audio","video","pdf"]},"inference._types__EmbeddingContentFormat":{"enum":["text","base64"]},"inference._types__RequestChatCompletion":{"type":"object","properties":{"messages":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__Message"},"description":"A list of objects representing the conversation.\nRequests should generally only add new messages from the user (role `user`).\nThe other message roles (`assistant`, `system`, or `tool`) should generally only be copied from the response to a previous completion request, such that the messages array is built up throughout a conversation."},"model":{"type":"string","description":"The ID of the model to use. By default, the model ID is set to the value included when creating the inference endpoint."},"max_completion_tokens":{"type":"number","description":"The upper bound limit for the number of tokens that can be generated for a completion request."},"reasoning":{"$ref":"./inference._types.json#/$defs/inference._types__Reasoning","description":"The reasoning configuration for the completion request.\nThis controls the model's reasoning process in one of two ways:\n\n* By specifying the model’s reasoning effort level with the `effort` field.\n* By enabling reasoning with default settings by setting `enabled` field to `true`.\n\nIt also includes optional settings to control:\n\n* The level of detail in the summary returned in the response with the `summary` field.\n* Whether reasoning details are included in the response at all with the `exclude` field.\n\nExample (effort):\n```\n{\n   \"reasoning\": {\n       \"effort\": \"high\",\n       \"summary\": \"concise\",\n       \"exclude\": false\n   }\n}\n```\nExample (enabled):\n```\n{\n   \"reasoning\": {\n       \"enabled\": true,\n       \"summary\": \"concise\",\n       \"exclude\": false\n   }\n}\n```\nCurrently supported only for `elastic` provider."},"stop":{"type":"array","items":{"type":"string"},"description":"A sequence of strings to control when the model should stop generating additional tokens."},"temperature":{"type":"number","description":"The sampling temperature to use."},"tool_choice":{"$ref":"./inference._types.json#/$defs/inference._types__CompletionToolType","description":"Controls which tool is called by the model.\nString representation: One of `auto`, `none`, or `requrired`. `auto` allows the model to choose between calling tools and generating a message. `none` causes the model to not call any tools. `required` forces the model to call one or more tools.\nExample (object representation):\n```\n{\n  \"tool_choice\": {\n      \"type\": \"function\",\n      \"function\": {\n          \"name\": \"get_current_weather\"\n      }\n  }\n}\n```"},"tools":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__CompletionTool"},"description":"A list of tools that the model can call.\nExample:\n```\n{\n  \"tools\": [\n      {\n          \"type\": \"function\",\n          \"function\": {\n              \"name\": \"get_price_of_item\",\n              \"description\": \"Get the current price of an item\",\n              \"parameters\": {\n                  \"type\": \"object\",\n                  \"properties\": {\n                      \"item\": {\n                          \"id\": \"12345\"\n                      },\n                      \"unit\": {\n                          \"type\": \"currency\"\n                      }\n                  }\n              }\n          }\n      }\n  ]\n}\n```"},"top_p":{"type":"number","description":"Nucleus sampling, an alternative to sampling with temperature."}},"required":["messages"]},"inference._types__Message":{"type":"object","properties":{"content":{"$ref":"./inference._types.json#/$defs/inference._types__MessageContent","description":"The content of the message.\n\nString example:\n```\n{\n   \"content\": \"Some string\"\n}\n```\n\nText example:\n```\n{\n  \"content\": [\n      {\n       \"text\": \"Some text\",\n       \"type\": \"text\"\n      }\n   ]\n}\n```\n\nImage example:\n```\n{\n  \"content\": [\n      {\n       \"image_url\": {\n         \"url\": \"data:image/jpeg;base64,...\"\n       },\n       \"type\": \"image_url\"\n      }\n   ]\n}\n```\n\nFile example:\n```\n{\n  \"content\": [\n      {\n       \"file\": {\n         \"file_data\": \"data:application/pdf;base64,...\",\n         \"filename\": \"somePDF\"\n       },\n       \"type\": \"file\"\n      }\n   ]\n}\n```"},"role":{"type":"string","description":"The role of the message author. Valid values are `user`, `assistant`, `system`, and `tool`."},"tool_call_id":{"type":"string","description":"Only for `tool` role messages. The tool call that this message is responding to."},"tool_calls":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__ToolCall"},"description":"Only for `assistant` role messages. The tool calls generated by the model. If it's specified, the `content` field is optional.\nExample:\n```\n{\n  \"tool_calls\": [\n      {\n          \"id\": \"call_KcAjWtAww20AihPHphUh46Gd\",\n          \"type\": \"function\",\n          \"function\": {\n              \"name\": \"get_current_weather\",\n              \"arguments\": \"{\\\"location\\\":\\\"Boston, MA\\\"}\"\n          }\n      }\n  ]\n}\n```"},"reasoning":{"type":"string","description":"Only for `assistant` role messages. The reasoning details generated by the model as plaintext.\nCurrently supported only for `elastic` provider."},"reasoning_details":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__ReasoningDetail"},"description":"Only for `assistant` role messages. The reasoning details generated by the model as structured data.\nCurrently supported only for `elastic` provider."}},"required":["role"],"description":"An object representing part of the conversation."},"inference._types__MessageContent":{"anyOf":[{"type":"string"},{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__ContentObject"}}]},"inference._types__ContentObject":{"type":"object","properties":{"type":{"$ref":"./inference._types.json#/$defs/inference._types__ContentType","description":"The type of content. Must be one of `text`, `image_url` or `file`. Not all services/models support content\ntypes other than \"text\""},"text":{"type":"string","description":"The text content. Only applicable for the `text` type"},"image_url":{"$ref":"./inference._types.json#/$defs/inference._types__ImageUrl","description":"The image content. Only applicable for the `image_url` type"},"file":{"$ref":"./inference._types.json#/$defs/inference._types__FileContent","description":"The file content. Only applicable for the `file` type"}},"required":["type","text","image_url","file"],"description":"An object style representation of a single portion of a conversation."},"inference._types__ContentType":{"enum":["text","image_url","file"]},"inference._types__ImageUrl":{"type":"object","properties":{"url":{"type":"string","description":"The base64 encoded image data as a data URI"},"detail":{"$ref":"./inference._types.json#/$defs/inference._types__ImageUrlDetail","description":"Specifies the detail level of the image"}},"required":["url"]},"inference._types__ImageUrlDetail":{"enum":["auto","low","high"]},"inference._types__FileContent":{"type":"object","properties":{"file_data":{"type":"string","description":"The base64 encoded file data"},"filename":{"type":"string","description":"The name of the file"}},"required":["file_data","filename"]},"inference._types__ToolCall":{"type":"object","properties":{"id":{"type":"string","description":"The identifier of the tool call."},"function":{"$ref":"./inference._types.json#/$defs/inference._types__ToolCallFunction","description":"The function that the model called."},"type":{"type":"string","description":"The type of the tool call."}},"required":["id","function","type"],"description":"A tool call generated by the model."},"inference._types__ToolCallFunction":{"type":"object","properties":{"arguments":{"type":"string","description":"The arguments to call the function with in JSON format."},"name":{"type":"string","description":"The name of the function to call."}},"required":["arguments","name"],"description":"The function that the model called."},"inference._types__ReasoningDetail":{"anyOf":[{"$ref":"./inference._types.json#/$defs/inference._types__EncryptedReasoningDetail"},{"$ref":"./inference._types.json#/$defs/inference._types__SummaryReasoningDetail"},{"$ref":"./inference._types.json#/$defs/inference._types__TextReasoningDetail"}],"description":"Type representing the different types of reasoning details that can be included in the response from the model.\nCurrently supported only for `elastic` provider."},"inference._types__EncryptedReasoningDetail":{"type":"object","properties":{"format":{"type":"string","description":"The format of the reasoning detail."},"id":{"type":"string","description":"The identifier of the reasoning detail."},"index":{"type":"number","description":"The index of the reasoning detail,\nwhich indicates its position in the sequence of reasoning details generated by the model."},"type":{"const":"reasoning.encrypted"},"data":{"type":"string","description":"The encrypted reasoning data generated by the model,\nwhich may be redacted or protected based on the model's configuration and the nature of the reasoning information."}},"required":["type","data"],"description":"The reasoning detail with encrypted reasoning data that may be redacted or protected."},"inference._types__SummaryReasoningDetail":{"type":"object","properties":{"format":{"type":"string","description":"The format of the reasoning detail."},"id":{"type":"string","description":"The identifier of the reasoning detail."},"index":{"type":"number","description":"The index of the reasoning detail,\nwhich indicates its position in the sequence of reasoning details generated by the model."},"type":{"const":"reasoning.summary"},"summary":{"type":"string","description":"The summary of the reasoning process generated by the model,\nwhich provides an overview of the key points and conclusions reached during the reasoning process."}},"required":["type","summary"],"description":"The reasoning summary detail includes a high-level summary of the model's reasoning process."},"inference._types__TextReasoningDetail":{"type":"object","properties":{"format":{"type":"string","description":"The format of the reasoning detail."},"id":{"type":"string","description":"The identifier of the reasoning detail."},"index":{"type":"number","description":"The index of the reasoning detail,\nwhich indicates its position in the sequence of reasoning details generated by the model."},"type":{"const":"reasoning.text"},"signature":{"type":"string","description":"The signature of the reasoning text,\nwhich can be used to verify the authenticity and integrity of the reasoning information provided by the model."},"text":{"type":"string","description":"The plaintext reasoning generated by the model,\nwhich provides a detailed explanation of the model's reasoning process in human-readable form."}},"required":["type"],"description":"The reasoning text detail includes plaintext reasoning with optional signature verification."},"inference._types__Reasoning":{"type":"object","properties":{"effort":{"$ref":"./inference._types.json#/$defs/inference._types__ReasoningEffort","description":"The level of effort the model should put into reasoning.\nThis is a hint that guides the model in how much effort to put into reasoning,\nwith `xhigh` being the most effort and `none` being no effort."},"enabled":{"type":"boolean","description":"Whether to enable reasoning with default settings.\nThis is a shortcut for enabling reasoning without having to specify the other parameters.\nIf `enabled` is set to `true`, then reasoning at the `medium` effort level is enabled.\nIgnored if `effort` is specified,\nin which case that parameter will control the reasoning process instead."},"exclude":{"type":"boolean","description":"Whether to exclude reasoning information from the response.\nIf `true`, the response will not include any reasoning details."},"summary":{"$ref":"./inference._types.json#/$defs/inference._types__ReasoningSummary","description":"The level of detail included in the reasoning summary returned in the response.\nThis is a hint on how much detail to include in the summary of the reasoning that is returned in the response,\nwith `auto` being the default level of detail, `concise` being less detail, and `detailed` being more detail."}},"description":"The reasoning configuration to use for the completion request.\nCurrently supported only for `elastic` provider."},"inference._types__ReasoningEffort":{"enum":["xhigh","high","medium","low","minimal","none"]},"inference._types__ReasoningSummary":{"enum":["auto","concise","detailed"]},"inference._types__CompletionToolType":{"anyOf":[{"type":"string"},{"$ref":"./inference._types.json#/$defs/inference._types__CompletionToolChoice"}]},"inference._types__CompletionToolChoice":{"type":"object","properties":{"type":{"type":"string","description":"The type of the tool."},"function":{"$ref":"./inference._types.json#/$defs/inference._types__CompletionToolChoiceFunction","description":"The tool choice function."}},"required":["type","function"],"description":"Controls which tool is called by the model."},"inference._types__CompletionToolChoiceFunction":{"type":"object","properties":{"name":{"type":"string","description":"The name of the function to call."}},"required":["name"],"description":"The tool choice function."},"inference._types__CompletionTool":{"type":"object","properties":{"type":{"type":"string","description":"The type of tool."},"function":{"$ref":"./inference._types.json#/$defs/inference._types__CompletionToolFunction","description":"The function definition."}},"required":["type","function"],"description":"A list of tools that the model can call."},"inference._types__CompletionToolFunction":{"type":"object","properties":{"description":{"type":"string","description":"A description of what the function does.\nThis is used by the model to choose when and how to call the function."},"name":{"type":"string","description":"The name of the function."},"parameters":{"description":"The parameters the functional accepts. This should be formatted as a JSON object."},"strict":{"type":"boolean","description":"Whether to enable schema adherence when generating the function call."}},"required":["name"],"description":"The completion tool function definition."},"inference._types__CompletionInferenceResult":{"type":"object","properties":{"completion":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__CompletionResult"}}},"required":["completion"],"description":"Defines the completion result."},"inference._types__CompletionResult":{"type":"object","properties":{"result":{"type":"string"}},"required":["result"],"description":"The completion result object"},"inference._types__TaskType":{"enum":["sparse_embedding","text_embedding","rerank","completion","chat_completion","embedding"]},"inference._types__DeleteInferenceEndpointResult":{"type":"object","properties":{"acknowledged":{"type":"boolean","description":"For a successful response, this value is always true. On failure, an exception is returned instead."},"pipelines":{"type":"array","items":{"type":"string"}}},"required":["acknowledged","pipelines"],"description":"Acknowledged response. For dry_run, contains the list of pipelines which reference the inference endpoint"},"inference._types__RequestEmbedding":{"type":"object","properties":{"input":{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingInput","description":"Inference input.\nEither a string, an array of strings, a `content` object, or an array of `content` objects.\n`content` objects may contain a single item or an array of items. Models that support multiple items per `content`\nobject will return a single embedding for each `content` object, regardless of how many items it contains.\nSupport for multiple items in a single `content` object is available in Elasticsearch 9.5.0 and later.\n\nstring example:\n```\n\"input\": \"Some text\"\n```\nstring array example:\n```\n\"input\": [\"Some text\", \"Some more text\"]\n```\n`content` object example:\n```\n\"input\": {\n    \"content\": {\n      \"type\": \"image\",\n      \"format\": \"base64\",\n      \"value\": \"data:image/jpeg;base64,...\"\n    }\n  }\n```\n`content` object array example:\n```\n\"input\": [\n  {\n    \"content\": {\n      \"type\": \"text\",\n      \"format\": \"text\",\n      \"value\": \"Some text to generate an embedding\"\n    }\n  },\n  {\n    \"content\": {\n      \"type\": \"image\",\n      \"format\": \"base64\",\n      \"value\": \"data:image/jpeg;base64,...\"\n    }\n  }\n]\n```\nMultiple items in one `content` object example (available in Elasticsearch 9.5.0 and later):\n```\n\"input\": [\n  {\n    \"content\": [\n      {\n        \"type\": \"image\",\n        \"format\": \"base64\",\n        \"value\": \"data:image/jpeg;base64,...\"\n      },\n      {\n        \"type\": \"text\",\n        \"value\": \"Some text to create an embedding\"\n      }\n    ]\n  }\n]\n```"},"input_type":{"type":"string","description":"The input data type for the embedding model. Possible values include:\n* `SEARCH`\n* `INGEST`\n* `CLASSIFICATION`\n* `CLUSTERING`\n\nNot all models support all values. Unsupported values will trigger a validation exception.\nAccepted values depend on the configured inference service, refer to the relevant service-specific documentation for more info.\n\n> info\n> The `input_type` parameter specified on the root level of the request body will take precedence over the `input_type` parameter specified in `task_settings`."},"task_settings":{"description":"Task settings for the individual inference request. These settings are specific to the <task_type> you specified and override the task settings specified when initializing the service."}},"required":["input"]},"inference._types__EmbeddingInput":{"anyOf":[{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingStringInput"},{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingContentInput"}],"description":"Inference input.\nEither a string, an array of strings, a `content` object, or an array of `content` objects."},"inference._types__EmbeddingStringInput":{"anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}}],"description":"Allows specifying text-only inputs for the `embedding` task."},"inference._types__EmbeddingContentInput":{"anyOf":[{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingContentObject"},{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingContentObject"}}],"description":"Allows specifying multimodal inputs for the `embedding` task."},"inference._types__EmbeddingContentObject":{"type":"object","properties":{"content":{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingContentObjectGroup","description":"An object or an array of objects containing the input data for the model to embed"}},"required":["content"],"description":"A wrapper object which contains the fields required to specify multimodal inputs"},"inference._types__EmbeddingContentObjectGroup":{"anyOf":[{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingContentObjectItem"},{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingContentObjectItem"}}],"description":"Allows specifying one or multiple items for the `embedding` task, which should result in a single embedding vector.\nSupport for multiple items is available in Elasticsearch 9.5.0 and later."},"inference._types__EmbeddingContentObjectItem":{"type":"object","properties":{"type":{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingContentType","description":"The type of input to embed. Not all models support all input types.\nThe `audio`, `video`, and `pdf` types are available in Elasticsearch 9.5.0 and later."},"format":{"$ref":"./inference._types.json#/$defs/inference._types__EmbeddingContentFormat","description":"The format of the input. For the `text` type this must be `text`. For all other types, this must be `base64`.\nIf not specified, this will default to `text` for the `text` type and `base64` for all other types."},"value":{"type":"string","description":"The value of the input to embed. For images, this must be a base64-encoded data URI, i.e. \"data:content/type;base64,...\""}},"required":["type","value"],"description":"An object containing the input data for a single item for the model to embed."},"inference._types__EmbeddingInferenceResult":{"type":"object","properties":{"embeddings_bytes":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"embeddings_bits":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"embeddings":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingResult"}}},"description":"EmbeddingInferenceResult is an aggregation of mutually exclusive embeddings variants"},"inference._types__DenseEmbeddingByteResult":{"type":"object","properties":{"embedding":{"$ref":"./inference._types.json#/$defs/inference._types__DenseByteVector"}},"required":["embedding"],"description":"The dense embedding result object for byte representation"},"inference._types__DenseByteVector":{"type":"array","items":{"type":"number"},"description":"Dense Embedding results containing bytes are represented as Dense\nVectors of bytes."},"inference._types__DenseEmbeddingResult":{"type":"object","properties":{"embedding":{"$ref":"./inference._types.json#/$defs/inference._types__DenseVector"}},"required":["embedding"],"description":"The dense embedding result object for float representation"},"inference._types__DenseVector":{"type":"array","items":{"type":"number"},"description":"Dense Embedding results are represented as Dense Vectors\nof floats."},"inference._types__InferenceEndpointInfo":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskType","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"],"description":"Represents an inference endpoint as returned by the GET API"},"inference._types__InferenceChunkingSettings":{"type":"object","properties":{"max_chunk_size":{"type":"number","description":"The maximum size of a chunk in words.\nThis value cannot be lower than `20` (for `sentence` strategy) or `10` (for `word` strategy).\nThis value should not exceed the window size for the associated model."},"overlap":{"type":"number","description":"The number of overlapping words for chunks.\nIt is applicable only to a `word` chunking strategy.\nThis value cannot be higher than half the `max_chunk_size` value."},"sentence_overlap":{"type":"number","description":"The number of overlapping sentences for chunks.\nIt is applicable only for a `sentence` chunking strategy.\nIt can be either `1` or `0`."},"separator_group":{"type":"string","description":"Only applicable to the `recursive` strategy and required when using it.\n\nSets a predefined list of separators in the saved chunking settings based on the selected text type.\nValues can be `markdown` or `plaintext`.\n\nUsing this parameter is an alternative to manually specifying a custom `separators` list."},"separators":{"type":"array","items":{"type":"string"},"description":"Only applicable to the `recursive` strategy and required when using it.\n\nA list of strings used as possible split points when chunking text.\n\nEach string can be a plain string or a regular expression (regex) pattern.\nThe system tries each separator in order to split the text, starting from the first item in the list.\n\nAfter splitting, it attempts to recombine smaller pieces into larger chunks that stay within\nthe `max_chunk_size` limit, to reduce the total number of chunks generated."},"strategy":{"type":"string","description":"The chunking strategy: `sentence`, `word`, `none` or `recursive`.\n\n * If `strategy` is set to `recursive`, you must also specify:\n\n- `max_chunk_size`\n- either `separators` or`separator_group`\n\nLearn more about different chunking strategies in the linked documentation."}},"description":"Chunking configuration object"},"inference._types__RegionPolicyDoc":{"type":"object","properties":{"region_policy":{"$ref":"./inference._types.json#/$defs/inference._types__RegionPolicy"},"created_at":{"$ref":"./_types.json#/$defs/_types__DateTime","description":"The date and time the region policy was created."},"created_by":{"type":"string","description":"The user who created the region policy."},"updated_at":{"$ref":"./_types.json#/$defs/_types__DateTime","description":"The date and time the region policy was last updated."},"updated_by":{"type":"string","description":"The user who last updated the region policy."}},"required":["region_policy","created_at"],"description":"The stored region policy document."},"inference._types__RegionPolicy":{"type":"object","properties":{"allowed_geos":{"type":"array","items":{"type":"string"},"description":"The list of allowed geographic areas.\nMutually exclusive with `allowed_regions`."},"allowed_regions":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__CspRegion"},"description":"The list of allowed cloud service provider regions.\nMutually exclusive with `allowed_geos`."}},"description":"The region policy configuration.\nSpecify exactly one of `allowed_geos` or `allowed_regions`."},"inference._types__CspRegion":{"type":"object","properties":{"csp":{"type":"string","description":"The cloud service provider, for example `aws`, `gcp`, or `azure`."},"region":{"type":"string","description":"The region of the cloud service provider, for example `us-east-1`."}},"required":["csp","region"],"description":"A cloud service provider region."},"inference._types__InferenceResult":{"type":"object","properties":{"embeddings_bytes":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"embeddings_bits":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"embeddings":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingResult"}},"text_embedding_bytes":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"text_embedding_bits":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"text_embedding":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingResult"}},"sparse_embedding":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__SparseEmbeddingResult"}},"completion":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__CompletionResult"}},"rerank":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__RankedDocument"}}},"description":"InferenceResult is an aggregation of mutually exclusive variants"},"inference._types__SparseEmbeddingResult":{"type":"object","properties":{"is_truncated":{"type":"boolean","description":"Indicates if the text input was truncated in the request sent to the service"},"embedding":{"$ref":"./inference._types.json#/$defs/inference._types__SparseVector"}},"required":["is_truncated","embedding"]},"inference._types__SparseVector":{"type":"object","additionalProperties":{"type":"number"},"description":"Sparse Embedding tokens are represented as a dictionary\nof string to double."},"inference._types__RankedDocument":{"type":"object","properties":{"index":{"type":"number"},"relevance_score":{"type":"number"},"text":{"type":"string"}},"required":["index","relevance_score"],"description":"The rerank result object representing a single ranked document\nid: the original index of the document in the request\nrelevance_score: the relevance_score of the document relative to the query\ntext: Optional, the text of the document, if requested"},"inference._types__InferenceEndpoint":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"}},"required":["service","service_settings"],"description":"Configuration options when storing the inference endpoint"},"inference._types__Ai21TaskType":{"enum":["completion","chat_completion"]},"inference._types__Ai21ServiceType":{"enum":["ai21"]},"inference._types__Ai21ServiceSettings":{"type":"object","properties":{"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the AI21 models documentation for the list of supported models and versions.\nService has been tested and confirmed to be working for `completion` and `chat_completion` tasks with the following models:\n* `jamba-mini`\n* `jamba-large`"},"api_key":{"type":"string","description":"A valid API key for accessing AI21 API.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the AI21 API.\nBy default, the `ai21` service sets the number of requests allowed per minute to 200. Please refer to AI21 documentation for more details."}},"required":["model_id"]},"inference._types__RateLimitSetting":{"type":"object","properties":{"requests_per_minute":{"type":"number","description":"The number of requests allowed per minute.\nBy default, the number of requests allowed per minute is set by each service as follows:\n\n* `alibabacloud-ai-search` service: `1000`\n* `amazonbedrock` service: `240`\n* `anthropic` service: `50`\n* `azureaistudio` service: `240`\n* `azureopenai` service and task type `text_embedding`: `1440`\n* `azureopenai` service and task types `completion` or `chat_completion`: `120`\n* `cohere` service: `10000`\n* `contextualai` service: `1000`\n* `elastic` service and task type `chat_completion`: `240`\n* `fireworksai` service: `6000`\n* `googleaistudio` service: `360`\n* `googlevertexai` service: `30000`\n* `hugging_face` service: `3000`\n* `jinaai` service: `2000`\n* `llama` service: `3000`\n* `mistral` service: `240`\n* `openai` service and task type `text_embedding`: `3000`\n* `openai` service and task type `completion`: `500`\n* `openshift_ai` service: `3000`\n* `voyageai` service: `2000`\n* `watsonxai` service: `120`"}},"description":"This setting helps to minimize the number of rate limit errors returned from the service."},"inference._types__InferenceEndpointInfoAi21":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAi21","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAi21":{"enum":["completion","chat_completion"]},"inference._types__AlibabaCloudTaskType":{"enum":["completion","rerank","sparse_embedding","text_embedding"]},"inference._types__AlibabaCloudServiceType":{"enum":["alibabacloud-ai-search"]},"inference._types__AlibabaCloudServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for the AlibabaCloud AI Search API."},"host":{"type":"string","description":"The name of the host address used for the inference task.\nYou can find the host address in the API keys section of the documentation."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from AlibabaCloud AI Search.\nBy default, the `alibabacloud-ai-search` service sets the number of requests allowed per minute to `1000`."},"service_id":{"type":"string","description":"The name of the model service to use for the inference task.\nThe following service IDs are available for the `completion` task:\n\n* `ops-qwen-turbo`\n* `qwen-turbo`\n* `qwen-plus`\n* `qwen-max ÷ qwen-max-longcontext`\n\nThe following service ID is available for the `rerank` task:\n\n* `ops-bge-reranker-larger`\n\nThe following service ID is available for the `sparse_embedding` task:\n\n* `ops-text-sparse-embedding-001`\n\nThe following service IDs are available for the `text_embedding` task:\n\n`ops-text-embedding-001`\n`ops-text-embedding-zh-001`\n`ops-text-embedding-en-001`\n`ops-text-embedding-002`"},"workspace":{"type":"string","description":"The name of the workspace used for the inference task."}},"required":["api_key","host","service_id","workspace"]},"inference._types__AlibabaCloudTaskSettings":{"type":"object","properties":{"input_type":{"type":"string","description":"For a `sparse_embedding` or `text_embedding` task, specify the type of input passed to the model.\nValid values are:\n\n* `ingest` for storing document embeddings in a vector database.\n* `search` for storing embeddings of search queries run against a vector database to find relevant documents."},"return_token":{"type":"boolean","description":"For a `sparse_embedding` task, it affects whether the token name will be returned in the response.\nIt defaults to `false`, which means only the token ID will be returned in the response."}}},"inference._types__InferenceEndpointInfoAlibabaCloudAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAlibabaCloudAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAlibabaCloudAI":{"enum":["text_embedding","rerank","completion","sparse_embedding"]},"inference._types__AmazonBedrockTaskType":{"enum":["chat_completion","completion","text_embedding"]},"inference._types__AmazonBedrockServiceType":{"enum":["amazonbedrock"]},"inference._types__AmazonBedrockServiceSettings":{"type":"object","properties":{"access_key":{"type":"string","description":"A valid AWS access key that has permissions to use Amazon Bedrock and access to models for inference requests."},"model":{"type":"string","description":"The base model ID or an ARN to a custom model based on a foundational model.\nThe base model IDs can be found in the Amazon Bedrock documentation.\nNote that the model ID must be available for the provider chosen and your IAM user must have access to the model."},"provider":{"type":"string","description":"The model provider for your deployment.\nNote that some providers may support only certain task types.\nSupported providers include:\n\n* `amazontitan` - available for `text_embedding` and `completion` task types\n* `anthropic` - available for `chat_completion` and `completion` task types\n* `ai21labs` - available for `chat_completion` and `completion` task types\n* `cohere` - available for `chat_completion`, `completion` and `text_embedding` task types\n* `meta` - available for `chat_completion` and `completion` task types\n* `mistral` - available for `chat_completion` and `completion` task types"},"region":{"type":"string","description":"The region that your model or ARN is deployed in.\nThe list of available regions per model can be found in the Amazon Bedrock documentation."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Amazon Bedrock.\nBy default, the `amazonbedrock` service sets the number of requests allowed per minute to 240."},"secret_key":{"type":"string","description":"A valid AWS secret key that is paired with the `access_key`.\nFor informationg about creating and managing access and secret keys, refer to the AWS documentation."}},"required":["access_key","model","region","secret_key"]},"inference._types__AmazonBedrockTaskSettings":{"type":"object","properties":{"max_new_tokens":{"type":"number","description":"For `chat_completion` and `completion` tasks, it sets the maximum number for the output tokens to be generated."},"temperature":{"type":"number","description":"For `chat_completion` and `completion` tasks, it is a number between 0.0 and 1.0 that controls the apparent creativity of the results.\nAt temperature 0.0 the model is most deterministic, at temperature 1.0 most random.\nIt should not be used if `top_p` or `top_k` is specified."},"top_k":{"type":"number","description":"For `chat_completion` and `completion` tasks, it limits samples to the top-K most likely words, balancing coherence and variability.\nIt is only available for anthropic, cohere, and mistral providers.\nIt is an alternative to `temperature`; it should not be used if `temperature` is specified."},"top_p":{"type":"number","description":"For `chat_completion` and `completion` tasks, it is a number in the range of 0.0 to 1.0, to eliminate low-probability tokens.\nTop-p uses nucleus sampling to select top tokens whose sum of likelihoods does not exceed a certain value, ensuring both variety and coherence.\nIt is an alternative to `temperature`; it should not be used if `temperature` is specified."}}},"inference._types__InferenceEndpointInfoAmazonBedrock":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAmazonBedrock","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAmazonBedrock":{"enum":["chat_completion","completion","text_embedding"]},"inference._types__TaskTypeAmazonSageMaker":{"enum":["text_embedding","completion","chat_completion","sparse_embedding","rerank"]},"inference._types__AmazonSageMakerServiceType":{"enum":["amazon_sagemaker"]},"inference._types__AmazonSageMakerServiceSettings":{"type":"object","properties":{"access_key":{"type":"string","description":"A valid AWS access key that has permissions to use Amazon SageMaker and access to models for invoking requests."},"endpoint_name":{"type":"string","description":"The name of the SageMaker endpoint."},"api":{"$ref":"./inference._types.json#/$defs/inference._types__AmazonSageMakerApi","description":"The API format to use when calling SageMaker.\nElasticsearch will convert the POST _inference request to this data format when invoking the SageMaker endpoint."},"region":{"type":"string","description":"The region that your endpoint or Amazon Resource Name (ARN) is deployed in.\nThe list of available regions per model can be found in the Amazon SageMaker documentation."},"secret_key":{"type":"string","description":"A valid AWS secret key that is paired with the `access_key`.\nFor information about creating and managing access and secret keys, refer to the AWS documentation."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__AmazonSageMakerSimilarity","description":"Required when `api` is `elastic` and task type is `text_embedding`. The similarity measure used when invoking the\n`text_embedding` task type."},"element_type":{"$ref":"./inference._types.json#/$defs/inference._types__AmazonSageMakerElementType","description":"Required when `api` is `elastic` and task type is `text_embedding`. The data type returned by the text embedding\nmodel.\nThis value is used when parsing the response back to Elasticsearch data structures."},"target_model":{"type":"string","description":"The model ID when calling a multi-model endpoint."},"target_container_hostname":{"type":"string","description":"The container to directly invoke when calling a multi-container endpoint."},"inference_component_name":{"type":"string","description":"The inference component to directly invoke when calling a multi-component endpoint."},"batch_size":{"type":"number","description":"The maximum number of inputs in each batch. This value is used by inference ingestion pipelines\nwhen processing semantic values. It correlates to the number of times the SageMaker endpoint is\ninvoked (one per batch of input)."},"dimensions":{"type":"number","description":"The number of dimensions returned by the text embedding models. If this value is not provided, then\nit is guessed by making invoking the endpoint for the `text_embedding` task."}},"required":["access_key","endpoint_name","api","region","secret_key"]},"inference._types__AmazonSageMakerApi":{"enum":["openai","elastic"]},"inference._types__AmazonSageMakerSimilarity":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__AmazonSageMakerElementType":{"enum":["byte","float","bit"]},"inference._types__AmazonSageMakerTaskSettings":{"type":"object","properties":{"custom_attributes":{"type":"string","description":"The AWS custom attributes passed verbatim through to the model running in the SageMaker Endpoint.\nValues will be returned in the `X-elastic-sagemaker-custom-attributes` header."},"enable_explanations":{"type":"string","description":"The optional JMESPath expression used to override the EnableExplanations provided during endpoint creation."},"inference_id":{"type":"string","description":"The capture data ID when enabled in the endpoint."},"session_id":{"type":"string","description":"The stateful session identifier for a new or existing session.\nNew sessions will be returned in the `X-elastic-sagemaker-new-session-id` header.\nClosed sessions will be returned in the `X-elastic-sagemaker-closed-session-id` header."},"target_variant":{"type":"string","description":"Specifies the variant when running with multi-variant Endpoints."}}},"inference._types__InferenceEndpointInfoAmazonSageMaker":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAmazonSageMaker","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__AnthropicTaskType":{"enum":["completion","chat_completion"]},"inference._types__AnthropicServiceType":{"enum":["anthropic"]},"inference._types__AnthropicServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for the Anthropic API."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Anthropic documentation for the list of supported models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Anthropic.\nBy default, the `anthropic` service sets the number of requests allowed per minute to 50."}},"required":["api_key","model_id"]},"inference._types__AnthropicTaskSettings":{"type":"object","properties":{"max_tokens":{"type":"number","description":"For a `completion` or `chat_completion` task, it is the maximum number of tokens to generate before stopping."},"temperature":{"type":"number","description":"For a `completion` or `chat_completion` task, it is the amount of randomness injected into the response.\nFor more details about the supported range, refer to Anthropic documentation."},"top_k":{"type":"number","description":"For a `completion` or `chat_completion` task, it specifies to only sample from the top K options for each subsequent token.\nIt is recommended for advanced use cases only.\nYou usually only need to use `temperature`."},"top_p":{"type":"number","description":"For a `completion` or `chat_completion` task, it specifies to use Anthropic's nucleus sampling.\nIn nucleus sampling, Anthropic computes the cumulative distribution over all the options for each subsequent token in decreasing probability order and cuts it off once it reaches the specified probability.\nYou should either alter `temperature` or `top_p`, but not both.\nIt is recommended for advanced use cases only.\nYou usually only need to use `temperature`."}},"required":["max_tokens"]},"inference._types__InferenceEndpointInfoAnthropic":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAnthropic","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAnthropic":{"enum":["completion","chat_completion"]},"inference._types__AzureAiStudioTaskType":{"enum":["completion","rerank","text_embedding"]},"inference._types__AzureAiStudioServiceType":{"enum":["azureaistudio"]},"inference._types__AzureAiStudioServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your Azure AI Studio model deployment.\nThis key can be found on the overview page for your deployment in the management section of your Azure AI Studio account.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"endpoint_type":{"type":"string","description":"The type of endpoint that is available for deployment through Azure AI Studio: `token` or `realtime`.\nThe `token` endpoint type is for \"pay as you go\" endpoints that are billed per token.\nThe `realtime` endpoint type is for \"real-time\" endpoints that are billed per hour of usage."},"target":{"type":"string","description":"The target URL of your Azure AI Studio model deployment.\nThis can be found on the overview page for your deployment in the management section of your Azure AI Studio account."},"provider":{"type":"string","description":"The model provider for your deployment.\nNote that some providers may support only certain task types.\nSupported providers include:\n\n* `cohere` - available for `text_embedding`, `rerank` and `completion` task types\n* `databricks` - available for `completion` task type only\n* `meta` - available for `completion` task type only\n* `microsoft_phi` - available for `completion` task type only\n* `mistral` - available for `completion` task type only\n* `openai` - available for `text_embedding` and `completion` task types"},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Azure AI Studio.\nBy default, the `azureaistudio` service sets the number of requests allowed per minute to 240."}},"required":["api_key","endpoint_type","target","provider"]},"inference._types__AzureAiStudioTaskSettings":{"type":"object","properties":{"do_sample":{"type":"number","description":"For a `completion` task, instruct the inference process to perform sampling.\nIt has no effect unless `temperature` or `top_p` is specified."},"max_new_tokens":{"type":"number","description":"For a `completion` task, provide a hint for the maximum number of output tokens to be generated."},"temperature":{"type":"number","description":"For a `completion` task, control the apparent creativity of generated completions with a sampling temperature.\nIt must be a number in the range of 0.0 to 2.0.\nIt should not be used if `top_p` is specified."},"top_p":{"type":"number","description":"For a `completion` task, make the model consider the results of the tokens with nucleus sampling probability.\nIt is an alternative value to `temperature` and must be a number in the range of 0.0 to 2.0.\nIt should not be used if `temperature` is specified."},"user":{"type":"string","description":"For a `text_embedding` task, specify the user issuing the request.\nThis information can be used for abuse detection."},"return_documents":{"type":"boolean","description":"For a `rerank` task, return doc text within the results."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents."}}},"inference._types__InferenceEndpointInfoAzureAIStudio":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAzureAIStudio","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAzureAIStudio":{"enum":["text_embedding","completion","rerank"]},"inference._types__AzureOpenAITaskType":{"enum":["completion","chat_completion","text_embedding"]},"inference._types__AzureOpenAIServiceType":{"enum":["azureopenai"]},"inference._types__AzureOpenAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Azure OpenAI account.\n\nIMPORTANT: You must specify either `api_key`, `entra_id`, or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint."},"api_version":{"type":"string","description":"The Azure API version ID to use.\nIt is recommended to use the latest supported non-preview version."},"client_id":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe application ID that's assigned to your app.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `tenant_id`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"client_secret":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe application secret that you created in the Microsoft app registration portal for your app.\n\nIMPORTANT: You must specify either `api_key`, `entra_id`, or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `tenant_id`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"deployment_id":{"type":"string","description":"The deployment name of your deployed models.\nYour Azure OpenAI deployments can be found though the Azure OpenAI Studio portal that is linked to your subscription."},"entra_id":{"type":"string","description":"A valid Microsoft Entra token.\n\nIMPORTANT: You must specify either `api_key`, `entra_id`, or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Azure.\nThe `azureopenai` service sets a default number of requests allowed per minute depending on the task type.\nFor `text_embedding`, it is set to `1440`.\nFor `completion` and `chat_completion`, it is set to `120`."},"resource_name":{"type":"string","description":"The name of your Azure OpenAI resource.\nYou can find this from the list of resources in the Azure Portal for your subscription."},"scopes":{"type":"array","items":{"type":"string"},"description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe resource identifier (application ID URI) of the resource you want, suffixed with .default\nFor example:\n```\n\"scopes\": [\n  \"https://cognitiveservices.azure.com/.default\"\n]\n```\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `tenant_id`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"tenant_id":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe directory tenant the application plans to operate against.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `tenant_id`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."}},"required":["api_version","deployment_id","resource_name"]},"inference._types__AzureOpenAITaskSettings":{"type":"object","properties":{"user":{"type":"string","description":"Specifies the user issuing the request.\nThis information can be used for abuse detection."},"headers":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies custom HTTP header parameters.\nFor example:\n```\n\"headers\": {\n  \"Custom-Header\": \"Some-Value\",\n  \"Another-Custom-Header\": \"Another-Value\"\n}\n```"}}},"inference._types__InferenceEndpointInfoAzureOpenAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeAzureOpenAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeAzureOpenAI":{"enum":["text_embedding","completion","chat_completion"]},"inference._types__CohereTaskType":{"enum":["completion","rerank","text_embedding"]},"inference._types__CohereServiceType":{"enum":["cohere"]},"inference._types__CohereServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Cohere account.\nYou can find or create your Cohere API keys on the Cohere API key settings page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"embedding_type":{"$ref":"./inference._types.json#/$defs/inference._types__CohereEmbeddingType","description":"For a `text_embedding` task, the types of embeddings you want to get back.\nUse `binary` for binary embeddings, which are encoded as bytes with signed int8 precision.\nUse `bit` for binary embeddings, which are encoded as bytes with signed int8 precision (this is a synonym of `binary`).\nUse `byte` for signed int8 embeddings (this is a synonym of `int8`).\nUse `float` for the default float embeddings.\nUse `int8` for signed int8 embeddings."},"model_id":{"type":"string","description":"For a `completion`, `rerank`, or `text_embedding` task, the name of the model to use for the inference task.\n\n* For the available `completion` models, refer to the [Cohere command docs](https://docs.cohere.com/docs/models#command).\n* For the available `rerank` models, refer to the [Cohere rerank docs](https://docs.cohere.com/reference/rerank-1).\n* For the available `text_embedding` models, refer to [Cohere embed docs](https://docs.cohere.com/reference/embed)."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Cohere.\nBy default, the `cohere` service sets the number of requests allowed per minute to 10000."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__CohereSimilarityType","description":"The similarity measure.\nIf the `embedding_type` is `float`, the default value is `dot_product`.\nIf the `embedding_type` is `int8` or `byte`, the default value is `cosine`."}},"required":["api_key","model_id"]},"inference._types__CohereEmbeddingType":{"enum":["binary","bit","byte","float","int8"]},"inference._types__CohereSimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__CohereTaskSettings":{"type":"object","properties":{"input_type":{"$ref":"./inference._types.json#/$defs/inference._types__CohereInputType","description":"For a `text_embedding` task, the type of input passed to the model.\nValid values are:\n\n* `classification`: Use it for embeddings passed through a text classifier.\n* `clustering`: Use it for the embeddings run through a clustering algorithm.\n* `ingest`: Use it for storing document embeddings in a vector database.\n* `search`: Use it for storing embeddings of search queries run against a vector database to find relevant documents.\n\nIMPORTANT: The `input_type` field is required when using embedding models `v3` and higher."},"return_documents":{"type":"boolean","description":"For a `rerank` task, return doc text within the results."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents.\nIf this inference endpoint is used in a `text_similarity_reranker` retriever query and `top_n` is set, it must be greater than or equal to `rank_window_size` in the query."},"truncate":{"$ref":"./inference._types.json#/$defs/inference._types__CohereTruncateType","description":"For a `text_embedding` task, the method to handle inputs longer than the maximum token length.\nValid values are:\n\n* `END`: When the input exceeds the maximum input token length, the end of the input is discarded.\n* `NONE`: When the input exceeds the maximum input token length, an error is returned.\n* `START`: When the input exceeds the maximum input token length, the start of the input is discarded."}},"required":["input_type"]},"inference._types__CohereInputType":{"enum":["classification","clustering","ingest","search"]},"inference._types__CohereTruncateType":{"enum":["END","NONE","START"]},"inference._types__InferenceEndpointInfoCohere":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeCohere","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeCohere":{"enum":["text_embedding","rerank","completion"]},"inference._types__TaskTypeContextualAI":{"enum":["rerank"]},"inference._types__ContextualAIServiceType":{"enum":["contextualai"]},"inference._types__ContextualAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Contexutual AI account.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Contextual AI documentation for the list of available rerank models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Contextual AI.\nThe `contextualai` service sets a default number of requests allowed per minute depending on the task type.\nFor `rerank`, it is set to `1000`."}},"required":["api_key","model_id"]},"inference._types__ContextualAITaskSettings":{"type":"object","properties":{"instruction":{"type":"string","description":"Instructions for the reranking model. Refer to <https://docs.contextual.ai/api-reference/rerank/rerank#body-instruction>\nOnly for the `rerank` task type."},"top_k":{"type":"number","description":"The number of most relevant documents to return.\nIf not specified, the reranking results of all documents will be returned.\nOnly for the `rerank` task type."}}},"inference._types__InferenceEndpointInfoContextualAi":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeContextualAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__CustomTaskType":{"enum":["text_embedding","sparse_embedding","rerank","completion"]},"inference._types__CustomServiceType":{"enum":["custom"]},"inference._types__CustomServiceSettings":{"type":"object","properties":{"batch_size":{"type":"number","description":"Specifies the batch size used for the semantic_text field. If the field is not provided, the default is 10.\nThe batch size is the maximum number of inputs in a single request to the upstream service.\nThe chunk within the batch are controlled by the selected chunking strategy for the semantic_text field."},"headers":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies the HTTP header parameters – such as `Authorization` or `Content-Type` – that are required to access the custom service.\nFor example:\n```\n\"headers\": {\n  \"Authorization\": \"Bearer ${api_key}\",\n  \"Content-Type\": \"application/json;charset=utf-8\"\n}\n```"},"input_type":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies the input type translation values that are used to replace the `${input_type}` template in the request body.\nFor example:\n```\n\"input_type\": {\n  \"translation\": {\n    \"ingest\": \"do_ingest\",\n    \"search\": \"do_search\"\n  },\n  \"default\": \"a_default\"\n},\n```\nIf the subsequent inference requests come from a search context, the `search` key will be used and the template will be replaced with `do_search`.\nIf it comes from the ingest context `do_ingest` is used. If it's a different context that is not specified, the default value will be used. If no default is specified an empty string is used.\n`translation` can be:\n* `classification`\n* `clustering`\n* `ingest`\n* `search`"},"query_parameters":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__CustomServiceQueryParameter"},"description":"Specifies the query parameters as a list of tuples. The arrays inside the `query_parameters` must have two items, a key and a value.\nFor example:\n```\n\"query_parameters\":[\n  [\"param_key\", \"some_value\"],\n  [\"param_key\", \"another_value\"],\n  [\"other_key\", \"other_value\"]\n]\n```\nIf the base url is `https://www.elastic.co` it results in: `https://www.elastic.co?param_key=some_value&param_key=another_value&other_key=other_value`."},"request":{"$ref":"./inference._types.json#/$defs/inference._types__CustomRequestParams","description":"The request configuration object."},"response":{"$ref":"./inference._types.json#/$defs/inference._types__CustomResponseParams","description":"The response configuration object."},"secret_parameters":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies secret parameters, like `api_key` or `api_token`, that are required to access the custom service.\nFor example:\n```\n\"secret_parameters\":{\n  \"api_key\":\"<api_key>\"\n}\n```"},"url":{"type":"string","description":"The URL endpoint to use for the requests."}},"required":["request","response","secret_parameters"]},"inference._types__CustomServiceQueryParameter":{"type":"array","items":{"type":"string"}},"inference._types__CustomRequestParams":{"type":"object","properties":{"content":{"type":"string","description":"The body structure of the request. It requires passing in the string-escaped result of the JSON format HTTP request body.\nFor example:\n```\n\"request\": \"{\\\"input\\\":${input}}\"\n```\n> info\n> The content string needs to be a single line except when using the Kibana console."}},"required":["content"]},"inference._types__CustomResponseParams":{"type":"object","properties":{"json_parser":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies the JSON parser that is used to parse the response from the custom service.\nDifferent task types require different json_parser parameters.\nFor example:\n```\n# text_embedding\n# For a response like this:\n\n{\n \"object\": \"list\",\n \"data\": [\n     {\n       \"object\": \"embedding\",\n       \"index\": 0,\n       \"embedding\": [\n           0.014539449,\n           -0.015288644\n       ]\n     }\n ],\n \"model\": \"text-embedding-ada-002-v2\",\n \"usage\": {\n     \"prompt_tokens\": 8,\n     \"total_tokens\": 8\n }\n}\n\n# the json_parser definition should look like this:\n\n\"response\":{\n  \"json_parser\":{\n    \"text_embeddings\":\"$.data[*].embedding[*]\"\n  }\n}\n\n# Elasticsearch supports the following embedding types:\n* float\n* byte\n* bit (or binary)\n\nTo specify the embedding type for the response, the `embedding_type`\nfield should be added in the `json_parser` object. Here's an example:\n\"response\":{\n  \"json_parser\":{\n    \"text_embeddings\":\"$.data[*].embedding[*]\",\n    \"embedding_type\":\"bit\"\n  }\n}\n\nIf `embedding_type` is not specified, it defaults to `float`.\n\n# sparse_embedding\n# For a response like this:\n\n{\n  \"request_id\": \"75C50B5B-E79E-4930-****-F48DBB392231\",\n  \"latency\": 22,\n  \"usage\": {\n     \"token_count\": 11\n  },\n  \"result\": {\n     \"sparse_embeddings\": [\n        {\n          \"index\": 0,\n          \"embedding\": [\n            {\n              \"token_id\": 6,\n              \"weight\": 0.101\n            },\n            {\n              \"token_id\": 163040,\n              \"weight\": 0.28417\n            }\n          ]\n        }\n     ]\n  }\n}\n\n# the json_parser definition should look like this:\n\n\"response\":{\n  \"json_parser\":{\n    \"token_path\":\"$.result.sparse_embeddings[*].embedding[*].token_id\",\n    \"weight_path\":\"$.result.sparse_embeddings[*].embedding[*].weight\"\n  }\n}\n\n# rerank\n# For a response like this:\n\n{\n  \"results\": [\n    {\n      \"index\": 3,\n      \"relevance_score\": 0.999071,\n      \"document\": \"abc\"\n    },\n    {\n      \"index\": 4,\n      \"relevance_score\": 0.7867867,\n      \"document\": \"123\"\n    },\n    {\n      \"index\": 0,\n      \"relevance_score\": 0.32713068,\n      \"document\": \"super\"\n    }\n  ],\n}\n\n# the json_parser definition should look like this:\n\n\"response\":{\n  \"json_parser\":{\n    \"reranked_index\":\"$.result.scores[*].index\",    // optional\n    \"relevance_score\":\"$.result.scores[*].score\",\n    \"document_text\":\"xxx\"    // optional\n  }\n}\n\n# completion\n# For a response like this:\n\n{\n \"id\": \"chatcmpl-B9MBs8CjcvOU2jLn4n570S5qMJKcT\",\n \"object\": \"chat.completion\",\n \"created\": 1741569952,\n \"model\": \"gpt-4.1-2025-04-14\",\n \"choices\": [\n   {\n    \"index\": 0,\n    \"message\": {\n      \"role\": \"assistant\",\n      \"content\": \"Hello! How can I assist you today?\",\n      \"refusal\": null,\n      \"annotations\": []\n    },\n    \"logprobs\": null,\n    \"finish_reason\": \"stop\"\n  }\n ]\n}\n\n# the json_parser definition should look like this:\n\n\"response\":{\n  \"json_parser\":{\n    \"completion_result\":\"$.choices[*].message.content\"\n  }\n}"}},"required":["json_parser"]},"inference._types__CustomTaskSettings":{"type":"object","properties":{"parameters":{"type":"object","additionalProperties":{"$ref":"./inference._types.json#/$defs/inference._types__CustomTaskParameter"},"description":"Specifies parameters that are required to run the custom service. The parameters depend on the model your custom service uses.\nFor example:\n```\n\"task_settings\":{\n  \"parameters\":{\n    \"input_type\":\"query\",\n    \"return_token\":true\n  }\n}\n```\n\n> warn\n> The `task_settings.parameters` cannot contain the same keys as `secret_parameters`. If they do, an error will be returned. This applies to PUT requests and POST requests."}}},"inference._types__CustomTaskParameter":{"anyOf":[{"type":"string"},{"type":"number"},{"type":"number"},{"type":"number"},{"type":"boolean"}]},"inference._types__InferenceEndpointInfoCustom":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeCustom","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeCustom":{"enum":["text_embedding","sparse_embedding","rerank","completion"]},"inference._types__TaskTypeDeepSeek":{"enum":["completion","chat_completion"]},"inference._types__DeepSeekServiceType":{"enum":["deepseek"]},"inference._types__DeepSeekServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your DeepSeek account.\nYou can find or create your DeepSeek API keys on the DeepSeek API key page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"model_id":{"type":"string","description":"For a `completion` or `chat_completion` task, the name of the model to use for the inference task.\n\nFor the available `completion` and `chat_completion` models, refer to the [DeepSeek Models & Pricing docs](https://api-docs.deepseek.com/quick_start/pricing)."},"url":{"type":"string","description":"The URL endpoint to use for the requests. Defaults to `https://api.deepseek.com/chat/completions`."}},"required":["api_key","model_id"]},"inference._types__InferenceEndpointInfoDeepSeek":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeDeepSeek","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__ElasticsearchTaskType":{"enum":["rerank","sparse_embedding","text_embedding"]},"inference._types__ElasticsearchServiceType":{"enum":["elasticsearch"]},"inference._types__ElasticsearchServiceSettings":{"type":"object","properties":{"adaptive_allocations":{"$ref":"./inference._types.json#/$defs/inference._types__AdaptiveAllocations","description":"Adaptive allocations configuration details.\nIf `enabled` is true, the number of allocations of the model is set based on the current load the process gets.\nWhen the load is high, a new model allocation is automatically created, respecting the value of `max_number_of_allocations` if it's set.\nWhen the load is low, a model allocation is automatically removed, respecting the value of `min_number_of_allocations` if it's set.\nIf `enabled` is true, do not set the number of allocations manually."},"deployment_id":{"type":"string","description":"The deployment identifier for a trained model deployment.\nWhen `deployment_id` is used the `model_id` is optional."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nIt can be the ID of a built-in model (for example, `.multilingual-e5-small` for E5) or a text embedding model that was uploaded by using the Eland client."},"num_allocations":{"type":"number","description":"The total number of allocations that are assigned to the model across machine learning nodes.\nIncreasing this value generally increases the throughput.\nIf adaptive allocations are enabled, do not set this value because it's automatically set."},"num_threads":{"type":"number","description":"The number of threads used by each model allocation during inference.\nThis setting generally increases the speed per inference request.\nThe inference process is a compute-bound process; `threads_per_allocations` must not exceed the number of available allocated processors per node.\nThe value must be a power of 2.\nThe maximum value is 32."},"long_document_strategy":{"type":"string","description":"Available only for the `rerank` task type using the Elastic reranker model.\nControls the strategy used for processing long documents during inference.\n\nPossible values:\n- `truncate` (default): Processes only the beginning of each document.\n- `chunk`: Splits long documents into smaller parts (chunks) before inference.\n\nWhen `long_document_strategy` is set to `chunk`, Elasticsearch splits each document into smaller parts but still returns a single score per document.\nThat score reflects the highest relevance score among all chunks."},"max_chunks_per_doc":{"type":"number","description":"Only for the `rerank` task type.\nLimits the number of chunks per document that are sent for inference when chunking is enabled.\nIf not set, all chunks generated for the document are processed."}},"required":["model_id","num_threads"]},"inference._types__AdaptiveAllocations":{"type":"object","properties":{"enabled":{"type":"boolean","description":"Turn on `adaptive_allocations`."},"max_number_of_allocations":{"type":"number","description":"The maximum number of allocations to scale to.\nIf set, it must be greater than or equal to `min_number_of_allocations`."},"min_number_of_allocations":{"type":"number","description":"The minimum number of allocations to scale to.\nIf set, it must be greater than or equal to 0.\nIf not defined, the deployment scales to 0."}}},"inference._types__ElasticsearchTaskSettings":{"type":"object","properties":{"return_documents":{"type":"boolean","description":"For a `rerank` task, return the document instead of only the index."}}},"inference._types__InferenceEndpointInfoElasticsearch":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeElasticsearch","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeElasticsearch":{"enum":["sparse_embedding","text_embedding","rerank"]},"inference._types__ElserTaskType":{"enum":["sparse_embedding"]},"inference._types__ElserServiceType":{"enum":["elser"]},"inference._types__ElserServiceSettings":{"type":"object","properties":{"adaptive_allocations":{"$ref":"./inference._types.json#/$defs/inference._types__AdaptiveAllocations","description":"Adaptive allocations configuration details.\nIf `enabled` is true, the number of allocations of the model is set based on the current load the process gets.\nWhen the load is high, a new model allocation is automatically created, respecting the value of `max_number_of_allocations` if it's set.\nWhen the load is low, a model allocation is automatically removed, respecting the value of `min_number_of_allocations` if it's set.\nIf `enabled` is true, do not set the number of allocations manually."},"num_allocations":{"type":"number","description":"The total number of allocations this model is assigned across machine learning nodes.\nIncreasing this value generally increases the throughput.\nIf adaptive allocations is enabled, do not set this value because it's automatically set."},"num_threads":{"type":"number","description":"The number of threads used by each model allocation during inference.\nIncreasing this value generally increases the speed per inference request.\nThe inference process is a compute-bound process; `threads_per_allocations` must not exceed the number of available allocated processors per node.\nThe value must be a power of 2.\nThe maximum value is 32.\n\n> info\n> If you want to optimize your ELSER endpoint for ingest, set the number of threads to 1. If you want to optimize your ELSER endpoint for search, set the number of threads to greater than 1."}},"required":["num_allocations","num_threads"]},"inference._types__InferenceEndpointInfoELSER":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeELSER","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeELSER":{"enum":["sparse_embedding"]},"inference._types__FireworksAITaskType":{"enum":["chat_completion","completion","text_embedding"]},"inference._types__FireworksAIServiceType":{"enum":["fireworksai"]},"inference._types__FireworksAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Fireworks AI account.\nYou can find or create your API keys in the Fireworks AI dashboard.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Fireworks AI documentation for the list of available models for chat completion, completion, and text embedding.\nFor text embedding, supported models include the Qwen3 embedding family (e.g. `fireworks/qwen3-embedding-8b`) and other models in the Fireworks model library."},"url":{"type":"string","description":"The URL endpoint to use for the requests.\nIf not provided, the default Fireworks AI API endpoint is used."},"dimensions":{"type":"number","description":"For a `text_embedding` task, the number of dimensions the resulting output embeddings should have.\nVariable-length embeddings are supported via this parameter."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__FireworksAISimilarityType","description":"For a `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Fireworks AI API.\nRate limit grouping is per API key only.\nBy default, the `fireworksai` service sets the number of requests allowed per minute to 6000."}},"required":["api_key","model_id"]},"inference._types__FireworksAISimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__FireworksAITaskSettings":{"type":"object","properties":{"user":{"type":"string","description":"For a `completion` or`chat_completion` task, specify the user issuing the request.\nThis information can be used for abuse detection."},"headers":{"type":"object","additionalProperties":{"type":"string"},"description":"For a `completion` or`chat_completion` task. Specifies custom HTTP header parameters.\nFor example:\n```\n\"headers\": {\n  \"Custom-Header\": \"Some-Value\",\n  \"Another-Custom-Header\": \"Another-Value\"\n}\n```"}}},"inference._types__InferenceEndpointInfoFireworksAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeFireworksAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeFireworksAI":{"enum":["chat_completion","completion","text_embedding"]},"inference._types__GoogleAiStudioTaskType":{"enum":["completion","text_embedding"]},"inference._types__GoogleAiServiceType":{"enum":["googleaistudio"]},"inference._types__GoogleAiStudioServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your Google Gemini account."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Google documentation for the list of supported models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Google AI Studio.\nBy default, the `googleaistudio` service sets the number of requests allowed per minute to 360."}},"required":["api_key","model_id"]},"inference._types__InferenceEndpointInfoGoogleAIStudio":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeGoogleAIStudio","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeGoogleAIStudio":{"enum":["text_embedding","completion"]},"inference._types__GoogleVertexAITaskType":{"enum":["rerank","text_embedding","completion","chat_completion"]},"inference._types__GoogleVertexAIServiceType":{"enum":["googlevertexai"]},"inference._types__GoogleVertexAIServiceSettings":{"type":"object","properties":{"provider":{"$ref":"./inference._types.json#/$defs/inference._types__GoogleModelGardenProvider","description":"The name of the Google Model Garden Provider for `completion` and `chat_completion` tasks.\nIn order for a Google Model Garden endpoint to be used `provider` must be defined and be other than `google`.\nModes:\n- Google Model Garden (third-party models): set `provider` to a supported non-`google` value and provide `url` and/or `streaming_url`.\n- Google Vertex AI: omit `provider` or set it to `google`. In this mode, do not set `url` or `streaming_url` and Elastic will construct the endpoint url from `location`, `model_id`, and `project_id` parameters."},"url":{"type":"string","description":"The URL for non-streaming `completion` requests to a Google Model Garden provider endpoint.\nIf both `url` and `streaming_url` are provided, each is used for its respective mode.\nIf `streaming_url` is not provided, `url` is also used for streaming `completion` and `chat_completion`.\nIf `provider` is not provided or set to `google` (Google Vertex AI), do not set `url` (or `streaming_url`).\nAt least one of `url` or `streaming_url` must be provided for Google Model Garden endpoint usage.\nCertain providers require separate URLs for streaming and non-streaming operations (e.g., Anthropic, Mistral, AI21). Others support both operation types through a single URL (e.g., Meta, Hugging Face).\nInformation on constructing the URL for various providers can be found in the Google Model Garden documentation for the model, or on the endpoint’s `Sample request` page. The request examples also illustrate the proper formatting for the `url`."},"streaming_url":{"type":"string","description":"The URL for streaming `completion` and `chat_completion` requests to a Google Model Garden provider endpoint.\nIf both `streaming_url` and `url` are provided, each is used for its respective mode.\nIf `url` is not provided, `streaming_url` is also used for non-streaming `completion` requests.\nIf `provider` is not provided or set to `google` (Google Vertex AI), do not set `streaming_url` (or `url`).\nAt least one of `streaming_url` or `url` must be provided for Google Model Garden endpoint usage.\nCertain providers require separate URLs for streaming and non-streaming operations (e.g., Anthropic, Mistral, AI21). Others support both operation types through a single URL (e.g., Meta, Hugging Face).\nInformation on constructing the URL for various providers can be found in the Google Model Garden documentation for the model, or on the endpoint’s `Sample request` page. The request examples also illustrate the proper formatting for the `streaming_url`."},"location":{"type":"string","description":"The name of the location to use for the inference task for the Google Vertex AI inference task.\nFor Google Vertex AI, when `provider` is omitted or `google`, `location` is optional.\nWhen `location` is omitted, the request targets the Google Vertex AI global endpoint (`https://aiplatform.googleapis.com`); when it is provided, the corresponding regional endpoint (`https://<location>-aiplatform.googleapis.com`) is used.\nThe field must be omitted to select the global endpoint; an empty string is not a valid `location`.\nFor Google Model Garden's `completion` and `chat_completion` tasks, when `provider` is a supported non-`google` value - `location` is ignored.\nRefer to the Google documentation for the list of supported locations."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nFor Google Vertex AI `model_id` is mandatory.\nFor Google Model Garden's `completion` and `chat_completion` tasks, when `provider` is a supported non-`google` value - `model_id` will be used for some providers that require it, otherwise - ignored.\nRefer to the Google documentation for the list of supported models for Google Vertex AI."},"project_id":{"type":"string","description":"The name of the project to use for the Google Vertex AI inference task.\nFor Google Vertex AI `project_id` is mandatory.\nFor Google Model Garden's `completion` and `chat_completion` tasks, when `provider` is a supported non-`google` value - `project_id` is ignored."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Google Vertex AI.\nBy default, the `googlevertexai` service sets the number of requests allowed per minute to 30.000."},"service_account_json":{"type":"string","description":"A valid service account in JSON format for the Google Vertex AI API."},"dimensions":{"type":"number","description":"For a `text_embedding` task, the number of dimensions the resulting output embeddings should have.\nBy default, the model's standard output dimension is used.\nRefer to the Google documentation for more information."},"max_batch_size":{"type":"number","description":"Only applicable for the `text_embedding` task type.\nControls the batch size of chunked inference requests sent to Google Vertex AI.\n\nSetting this parameter lower reduces the risk of exceeding token limits but may result in more API calls. Setting it higher increases throughput but may risk hitting token limits.\n\nTo estimate a safe `max_batch_size` value, you can use it together with the `max_chunk_size` parameter using the following formula:\n`max_batch_size ≈ max_chunk_size × 1.3 × 512 ÷ 20000`\n\nWhere:\n- `1.3` is an approximate tokens-per-word ratio\n- `512` is the maximum number of chunks that can be generated per document\n- `20000` is the Google Vertex AI token limit per request\n\nThis estimate assumes the worst-case scenario with a document generating the maximum 512 chunks."}},"required":["service_account_json"]},"inference._types__GoogleModelGardenProvider":{"enum":["google","anthropic","meta","hugging_face","mistral","ai21"]},"inference._types__GoogleVertexAITaskSettings":{"type":"object","properties":{"auto_truncate":{"type":"boolean","description":"For a `text_embedding` task, truncate inputs longer than the maximum token length automatically."},"top_n":{"type":"number","description":"For a `rerank` task, the number of the top N documents that should be returned."},"thinking_config":{"$ref":"./inference._types.json#/$defs/inference._types__ThinkingConfig","description":"For a `completion` or `chat_completion` task, allows configuration of the thinking features for the model.\nRefer to the Google documentation for the allowable configurations for each model type."},"max_tokens":{"type":"number","description":"For `completion` and `chat_completion` tasks, specifies the `max_tokens` value for requests sent to the Google Model Garden `anthropic` provider.\nIf `provider` is not set to `anthropic`, this field is ignored.\nIf `max_tokens` is specified - it must be a positive integer. If not specified, the default value of 1024 is used.\nAnthropic models require `max_tokens` to be set for each request. Please refer to the Anthropic documentation for more information."}}},"inference._types__ThinkingConfig":{"type":"object","properties":{"thinking_budget":{"type":"number","description":"Indicates the desired thinking budget in tokens."}}},"inference._types__InferenceEndpointInfoGoogleVertexAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeGoogleVertexAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeGoogleVertexAI":{"enum":["chat_completion","completion","text_embedding","rerank"]},"inference._types__GroqTaskType":{"enum":["chat_completion"]},"inference._types__GroqServiceType":{"enum":["groq"]},"inference._types__GroqServiceSettings":{"type":"object","properties":{"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Groq model documentation for the list of supported models and versions.\nService has been tested and confirmed to be working for `completion` and `chat_completion` tasks with the following models:\n* `llama-3.3-70b-versatile`"},"api_key":{"type":"string","description":"A valid API key for accessing Groq API.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Groq API.\nBy default, the `groq` service sets the number of requests allowed per minute to 1000. Refer to Groq documentation for more details."}},"required":["model_id"]},"inference._types__InferenceEndpointInfoGroq":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeGroq","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeGroq":{"enum":["chat_completion"]},"inference._types__HuggingFaceTaskType":{"enum":["chat_completion","completion","rerank","text_embedding"]},"inference._types__HuggingFaceServiceType":{"enum":["hugging_face"]},"inference._types__HuggingFaceServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid access token for your HuggingFace account.\nYou can create or find your access tokens on the HuggingFace settings page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Hugging Face.\nBy default, the `hugging_face` service sets the number of requests allowed per minute to 3000 for all supported tasks.\nHugging Face does not publish a universal rate limit — actual limits may vary.\nIt is recommended to adjust this value based on the capacity and limits of your specific deployment environment."},"url":{"type":"string","description":"The URL endpoint to use for the requests.\nFor `completion` and `chat_completion` tasks, the deployed model must be compatible with the Hugging Face Chat Completion interface (see the linked external documentation for details). The endpoint URL for the request must include `/v1/chat/completions`.\nIf the model supports the OpenAI Chat Completion schema, a toggle should appear in the interface. Enabling this toggle doesn't change any model behavior, it reveals the full endpoint URL needed (which should include `/v1/chat/completions`) when configuring the inference endpoint in Elasticsearch. If the model doesn't support this schema, the toggle may not be shown."},"model_id":{"type":"string","description":"The name of the HuggingFace model to use for the inference task.\nFor `completion` and `chat_completion` tasks, this field is optional but may be required for certain models — particularly when using serverless inference endpoints.\nFor the `text_embedding` task, this field should not be included. Otherwise, the request will fail."}},"required":["api_key","url"]},"inference._types__HuggingFaceTaskSettings":{"type":"object","properties":{"return_documents":{"type":"boolean","description":"For a `rerank` task, return doc text within the results."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents."}}},"inference._types__InferenceEndpointInfoHuggingFace":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeHuggingFace","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeHuggingFace":{"enum":["chat_completion","completion","rerank","text_embedding"]},"inference._types__JinaAITaskType":{"enum":["embedding","rerank","text_embedding"]},"inference._types__JinaAIServiceType":{"enum":["jinaai"]},"inference._types__JinaAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your JinaAI account.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"model_id":{"type":"string","description":"The name of the model to use for the inference task."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from JinaAI.\nBy default, the `jinaai` service sets the number of requests allowed per minute to 2000 for all task types."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__JinaAISimilarityType","description":"For an `embedding` or `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm.\nThe default values varies with the embedding type.\nFor example, a float embedding type uses a `dot_product` similarity measure by default."},"dimensions":{"type":"number","description":"For an `embedding` or `text_embedding` task, the number of dimensions the resulting output embeddings should have.\nBy default, the model's standard output dimension is used.\nRefer to the Jina documentation for more information."},"embedding_type":{"$ref":"./inference._types.json#/$defs/inference._types__JinaAIElementType","description":"For an `embedding` or `text_embedding` task, the data type returned by the model.\nUse `bit` for binary embeddings, which are encoded as bytes with signed int8 precision.\nUse `binary` for binary embeddings, which are encoded as bytes with signed int8 precision (this is a synonym of `bit`).\nUse `float` for the default float embeddings."},"multimodal_model":{"type":"boolean","description":"For the `embedding` task, whether the model supports multimodal inputs. If true, requests sent to the Jina model\nwill use the multimodal request format (a list of objects). If false, requests sent to the model will use the same\nformat as the `text_embedding` task (a list of strings). Setting this to `false` allows the `embedding` task to be\nused with models that do not support multimodal requests."}},"required":["api_key","model_id"]},"inference._types__JinaAISimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__JinaAIElementType":{"enum":["binary","bit","float"]},"inference._types__JinaAITaskSettings":{"type":"object","properties":{"return_documents":{"type":"boolean","description":"For a `rerank` task, return the doc text within the results."},"input_type":{"$ref":"./inference._types.json#/$defs/inference._types__JinaAITextEmbeddingTask","description":"For an `embedding` or `text_embedding` task, the task passed to the model.\nValid values are:\n\n* `classification`: Use it for embeddings passed through a classifier.\n* `clustering`: Use it for the embeddings run through a clustering algorithm.\n* `ingest`: Use it for storing document embeddings in a vector database.\n* `search`: Use it for storing embeddings of search queries run against a vector database to find relevant documents."},"late_chunking":{"type":"boolean","description":"For an `embedding` or `text_embedding` task, controls when text is split into chunks.\nWhen set to `true`, a request from Elasticsearch contains only chunks related to a single document. Instead of batching chunks across documents, Elasticsearch sends them in separate requests. This ensures that chunk embeddings retain context from the entire document, improving semantic quality.\n\nIf a document exceeds the model's context limits, or if the document contains non-text inputs (relevant when using the multimodal `embedding` task), late chunking is automatically disabled for that document only and standard chunking is used instead.\n\nIf not specified, defaults to `false`."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return.\nIt defaults to the number of the documents.\nIf this inference endpoint is used in a `text_similarity_reranker` retriever query and `top_n` is set, it must be greater than or equal to `rank_window_size` in the query."}}},"inference._types__JinaAITextEmbeddingTask":{"enum":["classification","clustering","ingest","search"]},"inference._types__InferenceEndpointInfoJinaAi":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeJinaAi","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeJinaAi":{"enum":["embedding","text_embedding","rerank"]},"inference._types__LlamaTaskType":{"enum":["text_embedding","completion","chat_completion"]},"inference._types__LlamaServiceType":{"enum":["llama"]},"inference._types__LlamaServiceSettings":{"type":"object","properties":{"url":{"type":"string","description":"The URL endpoint of the Llama stack endpoint.\nURL must contain:\n* For `text_embedding` task - `/v1/inference/embeddings`.\n* For `completion` and `chat_completion` tasks - `/v1/openai/v1/chat/completions`."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Llama downloading models documentation for different ways of getting a list of available models and downloading them.\nService has been tested and confirmed to be working with the following models:\n* For `text_embedding` task - `all-MiniLM-L6-v2`.\n* For `completion` and `chat_completion` tasks - `llama3.2:3b`."},"max_input_tokens":{"type":"number","description":"For a `text_embedding` task, the maximum number of tokens per input before chunking occurs."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__LlamaSimilarityType","description":"For a `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Llama API.\nBy default, the `llama` service sets the number of requests allowed per minute to 3000."}},"required":["url","model_id"]},"inference._types__LlamaSimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__InferenceEndpointInfoLlama":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeLlama","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeLlama":{"enum":["text_embedding","chat_completion","completion"]},"inference._types__MistralTaskType":{"enum":["text_embedding","completion","chat_completion"]},"inference._types__MistralServiceType":{"enum":["mistral"]},"inference._types__MistralServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your Mistral account.\nYou can find your Mistral API keys or you can create a new one on the API Keys page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"max_input_tokens":{"type":"number","description":"The maximum number of tokens per input before chunking occurs."},"model":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the Mistral models documentation for the list of available models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Mistral API.\nBy default, the `mistral` service sets the number of requests allowed per minute to 240."}},"required":["api_key","model"]},"inference._types__InferenceEndpointInfoMistral":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeMistral","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeMistral":{"enum":["text_embedding","chat_completion","completion"]},"inference._types__NvidiaTaskType":{"enum":["chat_completion","completion","rerank","text_embedding"]},"inference._types__NvidiaServiceType":{"enum":["nvidia"]},"inference._types__NvidiaServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your Nvidia endpoint.\nCan be found in `API Keys` section of Nvidia account settings."},"url":{"type":"string","description":"The URL of the Nvidia model endpoint. If not provided, the default endpoint URL is used depending on the task type:\n\n* For `text_embedding` task - `https://integrate.api.nvidia.com/v1/embeddings`.\n* For `completion` and `chat_completion` tasks - `https://integrate.api.nvidia.com/v1/chat/completions`.\n* For `rerank` task - `https://ai.api.nvidia.com/v1/retrieval/nvidia/reranking`."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the model's documentation for the name if needed.\nService has been tested and confirmed to be working with the following models:\n\n* For `text_embedding` task - `nvidia/llama-3.2-nv-embedqa-1b-v2`.\n* For `completion` and `chat_completion` tasks - `microsoft/phi-3-mini-128k-instruct`.\n* For `rerank` task - `nv-rerank-qa-mistral-4b:1`.\nService doesn't support `text_embedding` task `baai/bge-m3` and `nvidia/nvclip` models due to them not recognizing the `input_type` parameter."},"max_input_tokens":{"type":"number","description":"For a `text_embedding` task, the maximum number of tokens per input. Inputs exceeding this value are truncated prior to sending to the Nvidia API."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__NvidiaSimilarityType","description":"For a `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the Nvidia API.\nBy default, the `nvidia` service sets the number of requests allowed per minute to 3000."}},"required":["api_key","model_id"]},"inference._types__NvidiaSimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__NvidiaTaskSettings":{"type":"object","properties":{"input_type":{"$ref":"./inference._types.json#/$defs/inference._types__NvidiaInputType","description":"For a `text_embedding` task, type of input sent to the Nvidia endpoint.\nValid values are:\n\n* `ingest`: Mapped to Nvidia's `passage` value in request. Used when generating embeddings during indexing.\n* `search`: Mapped to Nvidia's `query` value in request. Used when generating embeddings during querying.\n\nIMPORTANT: For Nvidia endpoints, if the `input_type` field is not specified, it defaults to `query`."},"truncate":{"$ref":"./inference._types.json#/$defs/inference._types__CohereTruncateType","description":"For a `text_embedding` task, the method used by the Nvidia model to handle inputs longer than the maximum token length.\nValid values are:\n\n* `END`: When the input exceeds the maximum input token length, the end of the input is discarded.\n* `NONE`: When the input exceeds the maximum input token length, an error is returned.\n* `START`: When the input exceeds the maximum input token length, the start of the input is discarded."}}},"inference._types__NvidiaInputType":{"enum":["ingest","search"]},"inference._types__InferenceEndpointInfoNvidia":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference ID"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeNvidia","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeNvidia":{"enum":["chat_completion","completion","rerank","text_embedding"]},"inference._types__OpenAITaskType":{"enum":["chat_completion","completion","text_embedding","embedding"]},"inference._types__OpenAIServiceType":{"enum":["openai"]},"inference._types__OpenAIServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your OpenAI account.\nYou can find your OpenAI API keys in your OpenAI account under the API keys section.\n\nIMPORTANT: You must specify either `api_key` or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint."},"client_id":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe application ID that's assigned to your app.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `token_url`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"client_secret":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe application secret that you created for your app.\n\nIMPORTANT: You must specify either `api_key` or `client_secret`.\nIf you do not provide one or you provide more than one of them, you will receive an error when you try to create your endpoint.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `token_url`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"dimensions":{"type":"number","description":"For a `text_embedding` or `embedding` task, the number of dimensions the resulting output embeddings should have.\nIt is supported only in `text-embedding-3` and later models.\nIf it is not set, the OpenAI defined default for the model is used."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the OpenAI documentation for the list of available text embedding models."},"organization_id":{"type":"string","description":"The unique identifier for your organization.\nYou can find the Organization ID in your OpenAI account under *Settings > Organizations*."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from OpenAI.\nThe `openai` service sets a default number of requests allowed per minute depending on the task type.\nFor `text_embedding` and `embedding`, it is set to `3000`.\nFor `completion` and `chat_completion`, it is set to `500`."},"scopes":{"type":"array","items":{"type":"string"},"description":"For OAuth 2.0 authorization using the client credentials grant flow.\nThe resource identifier of the resource you want.\nFor example:\n```\n\"scopes\": [\n  \"scope1\",\n  \"scope2\"\n]\n```\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `token_url`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__OpenAISimilarityType","description":"For a `text_embedding` or `embedding` task, the similarity measure. One of `cosine`, `dot_product`, `l2_norm`. Defaults to `dot_product`."},"token_url":{"type":"string","description":"For OAuth 2.0 authorization using the client credentials grant flow.\nAn OAuth2 token endpoint where Elasticsearch sends a request to exchange client credentials for an access token.\n\nIMPORTANT: To configure OAuth 2.0, you must specify `client_id`, `scopes`, `token_url`, and `client_secret` together.\nIf one of the fields is missing, you will receive an error when you try to create your endpoint."},"url":{"type":"string","description":"The URL endpoint to use for the requests.\nIt can be changed for testing purposes.\nDefault value is `https://api.openai.com/v1/embeddings` for a `text_embedding` or `embedding` task,\n`https://api.openai.com/v1/chat/completions` for a `completion` or `chat_completion` task."}},"required":["model_id"]},"inference._types__OpenAISimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__OpenAITaskSettings":{"type":"object","properties":{"user":{"type":"string","description":"Specifies the user issuing the request.\nThis information can be used for abuse detection."},"headers":{"type":"object","additionalProperties":{"type":"string"},"description":"Specifies custom HTTP header parameters.\nFor example:\n```\n\"headers\": {\n  \"Custom-Header\": \"Some-Value\",\n  \"Another-Custom-Header\": \"Another-Value\"\n}\n```"}}},"inference._types__InferenceEndpointInfoOpenAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeOpenAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeOpenAI":{"enum":["text_embedding","chat_completion","completion","embedding"]},"inference._types__OpenShiftAiTaskType":{"enum":["text_embedding","completion","chat_completion","rerank"]},"inference._types__OpenShiftAiServiceType":{"enum":["openshift_ai"]},"inference._types__OpenShiftAiServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key for your OpenShift AI endpoint.\nCan be found in `Token authentication` section of model related information."},"url":{"type":"string","description":"The URL of the OpenShift AI hosted model endpoint."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the hosted model's documentation for the name if needed.\nService has been tested and confirmed to be working with the following models:\n* For `text_embedding` task - `gritlm-7b`.\n* For `completion` and `chat_completion` tasks - `llama-31-8b-instruct`.\n* For `rerank` task - `bge-reranker-v2-m3`."},"max_input_tokens":{"type":"number","description":"For a `text_embedding` task, the maximum number of tokens per input before chunking occurs."},"similarity":{"$ref":"./inference._types.json#/$defs/inference._types__OpenShiftAiSimilarityType","description":"For a `text_embedding` task, the similarity measure. One of cosine, dot_product, l2_norm.\nIf not specified, the default dot_product value is used."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from the OpenShift AI API.\nBy default, the `openshift_ai` service sets the number of requests allowed per minute to 3000."}},"required":["api_key","url"]},"inference._types__OpenShiftAiSimilarityType":{"enum":["cosine","dot_product","l2_norm"]},"inference._types__OpenShiftAiTaskSettings":{"type":"object","properties":{"return_documents":{"type":"boolean","description":"For a `rerank` task, whether to return the source documents in the response."},"top_n":{"type":"number","description":"For a `rerank` task, the number of most relevant documents to return."}}},"inference._types__InferenceEndpointInfoOpenShiftAi":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeOpenShiftAi","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeOpenShiftAi":{"enum":["text_embedding","chat_completion","completion","rerank"]},"inference._types__VoyageAITaskType":{"enum":["text_embedding","rerank"]},"inference._types__VoyageAIServiceType":{"enum":["voyageai"]},"inference._types__VoyageAIServiceSettings":{"type":"object","properties":{"dimensions":{"type":"number","description":"The number of dimensions for resulting output embeddings.\nThis setting maps to `output_dimension` in the VoyageAI documentation.\nOnly for the `text_embedding` task type."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the VoyageAI documentation for the list of available text embedding and rerank models."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from VoyageAI.\nThe `voyageai` service sets a default number of requests allowed per minute depending on the task type.\nFor both `text_embedding` and `rerank`, it is set to `2000`."},"embedding_type":{"type":"number","description":"The data type for the embeddings to be returned.\nThis setting maps to `output_dtype` in the VoyageAI documentation.\nPermitted values: float, int8, bit.\n`int8` is a synonym of `byte` in the VoyageAI documentation.\n`bit` is a synonym of `binary` in the VoyageAI documentation.\nOnly for the `text_embedding` task type."}},"required":["model_id"]},"inference._types__VoyageAITaskSettings":{"type":"object","properties":{"input_type":{"type":"string","description":"Type of the input text.\nPermitted values: `ingest` (maps to `document` in the VoyageAI documentation), `search` (maps to `query` in the VoyageAI documentation).\nOnly for the `text_embedding` task type."},"return_documents":{"type":"boolean","description":"Whether to return the source documents in the response.\nOnly for the `rerank` task type."},"top_k":{"type":"number","description":"The number of most relevant documents to return.\nIf not specified, the reranking results of all documents will be returned.\nOnly for the `rerank` task type."},"truncation":{"type":"boolean","description":"Whether to truncate the input texts to fit within the context length."}}},"inference._types__InferenceEndpointInfoVoyageAI":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeVoyageAI","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeVoyageAI":{"enum":["text_embedding","rerank"]},"inference._types__WatsonxTaskType":{"enum":["text_embedding","rerank","chat_completion","completion"]},"inference._types__WatsonxServiceType":{"enum":["watsonxai"]},"inference._types__WatsonxServiceSettings":{"type":"object","properties":{"api_key":{"type":"string","description":"A valid API key of your Watsonx account.\nYou can find your Watsonx API keys or you can create a new one on the API keys page.\n\nIMPORTANT: You need to provide the API key only once, during the inference model creation.\nThe get inference endpoint API does not retrieve your API key."},"api_version":{"type":"string","description":"A version parameter that takes a version date in the format of `YYYY-MM-DD`.\nFor the active version data parameters, refer to the Wastonx documentation."},"model_id":{"type":"string","description":"The name of the model to use for the inference task.\nRefer to the IBM Embedding Models section in the Watsonx documentation for the list of available text embedding models.\nRefer to the IBM library - Foundation models in Watsonx.ai."},"project_id":{"type":"string","description":"The identifier of the IBM Cloud project to use for the inference task."},"rate_limit":{"$ref":"./inference._types.json#/$defs/inference._types__RateLimitSetting","description":"This setting helps to minimize the number of rate limit errors returned from Watsonx.\nBy default, the `watsonxai` service sets the number of requests allowed per minute to 120."},"url":{"type":"string","description":"The URL of the inference endpoint that you created on Watsonx."}},"required":["api_key","api_version","model_id","project_id","url"]},"inference._types__InferenceEndpointInfoWatsonx":{"type":"object","properties":{"chunking_settings":{"$ref":"./inference._types.json#/$defs/inference._types__InferenceChunkingSettings","description":"The chunking configuration object.\nApplies only to the `embedding`, `sparse_embedding` and `text_embedding` task types.\nNot applicable to the `rerank`, `completion`, or `chat_completion` task types."},"service":{"type":"string","description":"The service type"},"service_settings":{"description":"Settings specific to the service"},"task_settings":{"description":"Task settings specific to the service and task type"},"inference_id":{"type":"string","description":"The inference Id"},"task_type":{"$ref":"./inference._types.json#/$defs/inference._types__TaskTypeWatsonx","description":"The task type"}},"required":["service","service_settings","inference_id","task_type"]},"inference._types__TaskTypeWatsonx":{"enum":["text_embedding","chat_completion","completion"]},"inference._types__RerankedInferenceResult":{"type":"object","properties":{"rerank":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__RankedDocument"}}},"required":["rerank"],"description":"Defines the response for a rerank request."},"inference._types__SparseEmbeddingInferenceResult":{"type":"object","properties":{"sparse_embedding":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__SparseEmbeddingResult"}}},"required":["sparse_embedding"],"description":"The response format for the sparse embedding request."},"inference._types__TextEmbeddingInferenceResult":{"type":"object","properties":{"text_embedding_bytes":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"text_embedding_bits":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingByteResult"}},"text_embedding":{"type":"array","items":{"$ref":"./inference._types.json#/$defs/inference._types__DenseEmbeddingResult"}}},"description":"TextEmbeddingInferenceResult is an aggregation of mutually exclusive text_embedding variants"}}}