gllm-inference-binary 0.4.40__cp312-cp312-manylinux_2_31_x86_64.whl → 0.5.6__cp312-cp312-manylinux_2_31_x86_64.whl

gllm_inference/builder/build_lm_invoker.pyi

@@ -65,7 +65,8 @@ def build_lm_invoker(model_id: str | ModelId, credentials: str | dict[str, Any]
             },
         )
         ```
-        Providing credentials through environment variable is not supported for Bedrock.
+        The credentials can also be provided through the `AWS_ACCESS_KEY_ID` and `AWS_SECRET_ACCESS_KEY`
+        environment variables.
 
         # Using Datasaur LLM Projects Deployment API
         ```python

gllm_inference/builder/build_lm_request_processor.pyi

@@ -1,7 +1,6 @@
 from _typeshed import Incomplete
 from gllm_inference.builder.build_lm_invoker import build_lm_invoker as build_lm_invoker
 from gllm_inference.builder.build_output_parser import build_output_parser as build_output_parser
-from gllm_inference.output_parser.output_parser import BaseOutputParser as BaseOutputParser
 from gllm_inference.prompt_builder import PromptBuilder as PromptBuilder
 from gllm_inference.request_processor.lm_request_processor import LMRequestProcessor as LMRequestProcessor
 from gllm_inference.schema.model_id import ModelId as ModelId
@@ -9,7 +8,7 @@ from typing import Any
 
 logger: Incomplete
 
-def build_lm_request_processor(model_id: str | ModelId, credentials: str | dict[str, Any] | None = None, config: dict[str, Any] | None = None, system_template: str = '', user_template: str = '', output_parser_type: str = 'none', output_parser: BaseOutputParser | None = None) -> LMRequestProcessor:
+def build_lm_request_processor(model_id: str | ModelId, credentials: str | dict[str, Any] | None = None, config: dict[str, Any] | None = None, system_template: str = '', user_template: str = '', key_defaults: dict[str, Any] | None = None, output_parser_type: str = 'none') -> LMRequestProcessor:
     '''Build a language model invoker based on the provided configurations.
 
     Args:
@@ -30,10 +29,11 @@ def build_lm_request_processor(model_id: str | ModelId, credentials: str | dict[
             Defaults to an empty string.
         user_template (str): The user prompt template. May contain placeholders enclosed in curly braces `{}`.
             Defaults to an empty string.
+        key_defaults (dict[str, str] | None, optional): Default values for the keys in the prompt templates.
+            Applied when the corresponding keys are not provided in the runtime input.
+            Defaults to None, in which case no default values will be assigned to the keys.
         output_parser_type (str, optional): The type of output parser to use. Supports "json" and "none".
             Defaults to "none".
-        output_parser (BaseOutputParser | None, optional): Deprecated parameter to pass an output parser.
-            Will be removed in v0.5.0. Defaults to None.
 
     Returns:
         LMRequestProcessor: The initialized language model request processor.
@@ -66,13 +66,14 @@ def build_lm_request_processor(model_id: str | ModelId, credentials: str | dict[
         )
         ```
 
-        # With system template
+        # With custom prompt builder configuration
         ```python
         lm_request_processor = build_lm_request_processor(
             model_id="openai/gpt-4o-mini",
             credentials="sk-...",
-            system_template="Talk like a pirate.",
+            system_template="Talk like a {role}.",
             user_template="{query}",
+            key_defaults={"role": "pirate"},
         )
         ```
 

gllm_inference/catalog/catalog.pyi

@@ -13,46 +13,42 @@ class BaseCatalog(ABC, BaseModel, Generic[T], arbitrary_types_allowed=True, meta
     Attributes:
         components (dict[str, T]): A dictionary containing the components.
 
-    Load from Google Sheets using client email and private key example:
-    ```python
-    catalog = BaseCatalog.from_gsheets(
-        sheet_id="...",
-        worksheet_id="...",
-        client_email="...",
-        private_key="...",
-    )
-
-    component = catalog.name
-    ```
-
-    Load from Google Sheets using credential file example:
-    ```python
-    catalog = BaseCatalog.from_gsheets(
-        sheet_id="...",
-        worksheet_id="...",
-        credential_file_path="...",
-    )
-
-    component = catalog.name
-    ```
-
-    Load from CSV example:
-    ```python
-    catalog = BaseCatalog.from_csv(csv_path="...")
-
-    component = catalog.name
-    ```
-
-    Load from records example:
-    ```python
-    catalog = BaseCatalog.from_records(
-        records=[
+    Initialization:
+        # Example 1: Load from Google Sheets using client email and private key
+        ```python
+        catalog = BaseCatalog.from_gsheets(
+            sheet_id="...",
+            worksheet_id="...",
+            client_email="...",
+            private_key="...",
+        )
+        component = catalog.name
+        ```
+
+        # Example 2: Load from Google Sheets using credential file
+        ```python
+        catalog = BaseCatalog.from_gsheets(
+            sheet_id="...",
+            worksheet_id="...",
+            credential_file_path="...",
+        )
+        component = catalog.name
+        ```
+
+        # Example 3: Load from CSV
+        ```python
+        catalog = BaseCatalog.from_csv(csv_path="...")
+        component = catalog.name
+        ```
+
+        # Example 4: Load from records
+        ```python
+        records = [
             {"name": "...", "col_1": "...", "col_2": "..."},
             {"name": "...", "col_1": "...", "col_2": "..."},
-        ],
-    )
-
-    component = catalog.name
+        ]
+        catalog = BaseCatalog.from_records(records=records)
+        component = catalog.name
     ```
     '''
     components: dict[str, T]

gllm_inference/catalog/lm_request_processor_catalog.pyi

@@ -1,15 +1,12 @@
 from _typeshed import Incomplete
-from gllm_inference.builder import build_lm_request_processor as build_lm_request_processor, build_output_parser as build_output_parser
-from gllm_inference.catalog.catalog import BaseCatalog as BaseCatalog, logger as logger
-from gllm_inference.catalog.component_map import LM_INVOKER_TYPE_MAP as LM_INVOKER_TYPE_MAP, PROMPT_BUILDER_TYPE_MAP as PROMPT_BUILDER_TYPE_MAP
-from gllm_inference.lm_invoker.lm_invoker import BaseLMInvoker as BaseLMInvoker
-from gllm_inference.multimodal_lm_invoker.multimodal_lm_invoker import BaseMultimodalLMInvoker as BaseMultimodalLMInvoker
-from gllm_inference.multimodal_prompt_builder.multimodal_prompt_builder import MultimodalPromptBuilder as MultimodalPromptBuilder
-from gllm_inference.output_parser.output_parser import BaseOutputParser as BaseOutputParser
-from gllm_inference.prompt_builder.prompt_builder import BasePromptBuilder as BasePromptBuilder, PromptBuilder as PromptBuilder
+from gllm_inference.builder import build_lm_request_processor as build_lm_request_processor
+from gllm_inference.catalog.catalog import BaseCatalog as BaseCatalog
 from gllm_inference.request_processor import LMRequestProcessor as LMRequestProcessor
 
+MODEL_ID_ENV_VAR_REGEX_PATTERN: str
 LM_REQUEST_PROCESSOR_REQUIRED_COLUMNS: Incomplete
+CONFIG_SCHEMA_MAP: Incomplete
+logger: Incomplete
 
 class LMRequestProcessorCatalog(BaseCatalog[LMRequestProcessor]):
     '''Loads multiple LM request processors from certain sources.
@@ -50,49 +47,66 @@ class LMRequestProcessorCatalog(BaseCatalog[LMRequestProcessor]):
 
         # Example 4: Load from record
         ```python
-        catalog = LMRequestProcessorCatalog.from_records(
-            name="...",
-            system_template="...",
-            user_template="...",
-            model_id="...",
-            credentials="...",
-            config="...",
-            output_parser_type="...",
-        )
-
-        lm_request_processor = catalog.name
+        records=[
+            {
+                "name": "answer_question",
+                "system_template": (
+                    "You are helpful assistant.\\n"
+                    "Answer the following question based on the provided context.\\n"
+                    "```{context}```"
+                ),
+                "user_template": "{query}",
+                "key_defaults": \'{"context": "<default context>"}\',
+                "model_id": "openai/gpt-4.1-nano",
+                "credentials": "OPENAI_API_KEY",
+                "config": "",
+                "output_parser_type": "none",
+            },
+        ]
+        catalog = LMRequestProcessorCatalog.from_records(records=records)
+        lm_request_processor = catalog.answer_question
         ```
 
     Template Format Example:
         # Example 1: Google Sheets
-        For an example of how a Google Sheets file can be formatted to be loaded using LMRequestProcessorCatalog,
-        see: https://docs.google.com/spreadsheets/d/1CX9i45yEinv1UdB3s6uHNMj7mxr2-s1NFHfFDvMsq0E/edit?usp=drive_link
+        For an example of how a Google Sheets file can be formatted to be loaded using LMRequestProcessorCatalog, see:
+        https://docs.google.com/spreadsheets/d/1CX9i45yEinv1UdB3s6uHNMj7mxr2-s1NFHfFDvMsq0E/edit?usp=drive_link
 
         # Example 2: CSV
         For an example of how a CSV file can be formatted to be loaded using LMRequestProcessorCatalog, see:
-        https://drive.google.com/file/d/11FzAy_7IoKFhHSKVcC1HFMBbWIyWtbcT/view?usp=drive_link
+        https://drive.google.com/file/d/1_2rSoxh3CR2KZxIyUmpowMrt0Lm0YqAb/view?usp=drive_link
 
     Template Explanation:
         The required columns are:
-            1. name (str): The name of the LM request processor.
-            2. system_template (str): The system template of the prompt builder.
-            3. user_template (str): The user template of the prompt builder.
-            4. model_id (str): The model ID of the LM invoker.
-            5. credentials (str | json_str): The credentials of the LM invoker.
-            6. config (json_str): The additional configuration of the LM invoker.
-            7. output_parser_type (str): The type of the output parser.
+        1. name (str): The name of the LM request processor.
+        2. system_template (str): The system template of the prompt builder.
+        3. user_template (str): The user template of the prompt builder.
+        4. key_defaults (json_str): The default values for the prompt template keys.
+        5. model_id (str): The model ID of the LM invoker.
+        6. credentials (str | json_str): The credentials of the LM invoker.
+        7. config (json_str): The additional configuration of the LM invoker.
+        8. output_parser_type (str): The type of the output parser.
 
         Important Notes:
         1. At least one of `system_template` or `user_template` must be filled.
-        2. `model_id` must be filled with the model ID of the LM invoker. For available model IDs, see:
-            https://gdplabs.gitbook.io/sdk/supported-models
-        3. `credentials` is optional. If it is filled, it can either be:
-            3.1. An environment variable name containing the API key (e.g. OPENAI_API_KEY).
-            3.2. A path to a credentials JSON file, currently only supported for Google Vertex AI.
-            3.3. A dictionary of credentials, currently supported for Bedrock and LangChain.
-        4. `config` is optional. When this column is empty, the LM invoker will use the
-            default configuration. If it is filled, it must be a valid JSON string.
-        5. `output_parser_type` can either be:
-            5.1. none: No output parser will be used.
-            5.2. json: The JSONOutputParser will be used.
+        2. `key_defaults` is optional. If filled, must be a dictionary containing the default values for the
+            prompt template keys. These default values will be applied when the corresponding keys are not provided
+            in the runtime input. If it is empty, the prompt template keys will not have default values.
+        3. The `model_id`:
+            3.1. Must be filled with the model ID of the LM invoker, e.g. "openai/gpt-4.1-nano".
+            3.2. Can be partially loaded from the environment variable using the "${ENV_VAR_KEY}" syntax,
+                e.g. "azure-openai/${AZURE_ENDPOINT}/${AZURE_DEPLOYMENT}".
+            3.3. For the available model ID formats, see: https://gdplabs.gitbook.io/sdk/resources/supported-models
+        4. `credentials` is optional. If it is filled, it can either be:
+            4.1. An environment variable name containing the API key (e.g. OPENAI_API_KEY).
+            4.2. An environment variable name containing the path to a credentials JSON file
+                (e.g. GOOGLE_CREDENTIALS_FILE_PATH). Currently only supported for Google Vertex AI.
+            4.3. A dictionary of credentials, with each value being an environment variable name corresponding to the
+                credential (e.g. {"api_key": "OPENAI_API_KEY"}). Currently supported for Bedrock and LangChain.
+            If it is empty, the LM invoker will use the default credentials loaded from the environment variables.
+        5. `config` is optional. If filled, must be a dictionary containing the configuration for the LM invoker.
+            If it is empty, the LM invoker will use the default configuration.
+        6. `output_parser_type` can either be:
+            6.1. none: No output parser will be used.
+            6.2. json: The JSONOutputParser will be used.
     '''

gllm_inference/catalog/prompt_builder_catalog.pyi

@@ -1,18 +1,15 @@
 from _typeshed import Incomplete
-from gllm_inference.catalog.catalog import BaseCatalog as BaseCatalog, logger as logger
-from gllm_inference.catalog.component_map import PROMPT_BUILDER_TYPE_MAP as PROMPT_BUILDER_TYPE_MAP
-from gllm_inference.multimodal_prompt_builder import MultimodalPromptBuilder as MultimodalPromptBuilder
-from gllm_inference.prompt_builder.prompt_builder import BasePromptBuilder as BasePromptBuilder, PromptBuilder as PromptBuilder
+from gllm_inference.catalog.catalog import BaseCatalog as BaseCatalog
+from gllm_inference.prompt_builder.prompt_builder import PromptBuilder as PromptBuilder
 
-PROMPT_BUILDER_MODEL_PARAM_MAP: Incomplete
 PROMPT_BUILDER_REQUIRED_COLUMNS: Incomplete
+logger: Incomplete
 
-class PromptBuilderCatalog(BaseCatalog[BasePromptBuilder | MultimodalPromptBuilder | PromptBuilder]):
+class PromptBuilderCatalog(BaseCatalog[PromptBuilder]):
     '''Loads multiple prompt builders from certain sources.
 
     Attributes:
-        components (dict[str, BasePromptBuilder | MultimodalPromptBuilder | PromptBuilder]):
-            Dictionary of the loaded prompt builders.
+        components (dict[str, PromptBuilder]): Dictionary of the loaded prompt builders.
 
     Initialization:
         # Example 1: Load from Google Sheets using client email and private key
@@ -23,7 +20,6 @@ class PromptBuilderCatalog(BaseCatalog[BasePromptBuilder | MultimodalPromptBuild
             client_email="...",
             private_key="...",
         )
-
         prompt_builder = catalog.name
         ```
 
@@ -34,42 +30,30 @@ class PromptBuilderCatalog(BaseCatalog[BasePromptBuilder | MultimodalPromptBuild
             worksheet_id="...",
             credential_file_path="...",
         )
-
         prompt_builder = catalog.name
         ```
 
         # Example 3: Load from CSV
         ```python
         catalog = PromptBuilderCatalog.from_csv(csv_path="...")
-
         prompt_builder = catalog.name
         ```
 
         # Example 4: Load from records
         ```python
-        catalog = PromptBuilderCatalog.from_records(
-            records=[
-                {
-                    "name": "summarize",
-                    "system": "You are an AI expert\\nSummarize the following context.\\n\\nContext:\\n```{context}```",
-                    "user": ""
-                },
-                {
-                    "name": "transform_query",
-                    "system": "",
-                    "user": "Transform the following query into a simpler form.\\n\\nQuery:\\n```{query}```"
-                },
-                {
-                    "name": "draft_document",
-                    "system": (
-                        "You are an AI expert.\\nDraft a document following the provided format and context.\\n\\n"
-                        "Format:\\n```{format}```\\n\\nContext:\\n```{context}```"
-                    ),
-                    "user": "User instruction:\\n{query}"
-                },
-            ]
-        )
-
+        records=[
+            {
+                "name": "answer_question",
+                "system": (
+                    "You are helpful assistant.\\n"
+                    "Answer the following question based on the provided context.\\n"
+                    "```{context}```"
+                ),
+                "user": "{query}",
+                "key_defaults": \'{"context": "<default context>"}\',
+            },
+        ]
+        catalog = PromptBuilderCatalog.from_records(records=records)
         prompt_builder = catalog.answer_question
         ```
 
@@ -80,18 +64,19 @@ class PromptBuilderCatalog(BaseCatalog[BasePromptBuilder | MultimodalPromptBuild
 
         # Example 2: CSV
         For an example of how a CSV file can be formatted to be loaded using PromptBuilderCatalog, see:
-        https://drive.google.com/file/d/1CWijOk-g16ZglUn_K2bDPmbyyBDK2r0L/view?usp=drive_link
+        https://drive.google.com/file/d/1KQgddMdbcZBZmroQFtjSl-TKLohq84Fz/view?usp=drive_link
 
 
     Template explanation:
         The required columns are:
-            1. name (str): The name of the prompt builder.
-            2. system (str): The system template of the prompt builder.
-            3. user (str): The user template of the prompt builder.
+        1. name (str): The name of the prompt builder.
+        2. system (str): The system template of the prompt builder.
+        3. user (str): The user template of the prompt builder.
+        4. key_defaults (json_str): The default values for the prompt template keys.
 
         Important Notes:
-            1. At least one of the `system` and `user` columns must be filled.
-
-    WARNING: The use of BasePromptBuilder | MultimodalPromptBuilder is deprecated and will be removed in version 0.5.0.
-    Please use PromptBuilder instead.
+        1. At least one of the `system` and `user` columns must be filled.
+        2. `key_defaults` is optional. If filled, must be a dictionary containing the default values for the
+            prompt template keys. These default values will be applied when the corresponding keys are not provided
+            in the runtime input. If it is empty, the prompt template keys will not have default values.
     '''

gllm_inference/constants.pyi

@@ -1,10 +1,8 @@
 from _typeshed import Incomplete
 
-ALL_EXTENSIONS: str
 DEFAULT_AZURE_OPENAI_API_VERSION: str
 DOCUMENT_MIME_TYPES: Incomplete
 GOOGLE_SCOPES: Incomplete
+INVOKER_DEFAULT_TIMEOUT: float
 HEX_REPR_LENGTH: int
-MESSAGE_TUPLE_LENGTH: int
-DEFAULT_CONTENT_PLACEHOLDER_TYPE: str
 HTTP_STATUS_CODE_PATTERNS: Incomplete

gllm_inference/em_invoker/__init__.pyi

@@ -1,12 +1,9 @@
 from gllm_inference.em_invoker.azure_openai_em_invoker import AzureOpenAIEMInvoker as AzureOpenAIEMInvoker
 from gllm_inference.em_invoker.google_em_invoker import GoogleEMInvoker as GoogleEMInvoker
-from gllm_inference.em_invoker.google_generativeai_em_invoker import GoogleGenerativeAIEMInvoker as GoogleGenerativeAIEMInvoker
-from gllm_inference.em_invoker.google_vertexai_em_invoker import GoogleVertexAIEMInvoker as GoogleVertexAIEMInvoker
 from gllm_inference.em_invoker.langchain_em_invoker import LangChainEMInvoker as LangChainEMInvoker
 from gllm_inference.em_invoker.openai_compatible_em_invoker import OpenAICompatibleEMInvoker as OpenAICompatibleEMInvoker
 from gllm_inference.em_invoker.openai_em_invoker import OpenAIEMInvoker as OpenAIEMInvoker
-from gllm_inference.em_invoker.tei_em_invoker import TEIEMInvoker as TEIEMInvoker
 from gllm_inference.em_invoker.twelevelabs_em_invoker import TwelveLabsEMInvoker as TwelveLabsEMInvoker
 from gllm_inference.em_invoker.voyage_em_invoker import VoyageEMInvoker as VoyageEMInvoker
 
-__all__ = ['AzureOpenAIEMInvoker', 'GoogleEMInvoker', 'GoogleGenerativeAIEMInvoker', 'GoogleVertexAIEMInvoker', 'LangChainEMInvoker', 'OpenAIEMInvoker', 'OpenAICompatibleEMInvoker', 'TEIEMInvoker', 'TwelveLabsEMInvoker', 'VoyageEMInvoker']
+__all__ = ['AzureOpenAIEMInvoker', 'GoogleEMInvoker', 'LangChainEMInvoker', 'OpenAIEMInvoker', 'OpenAICompatibleEMInvoker', 'TwelveLabsEMInvoker', 'VoyageEMInvoker']

gllm_inference/em_invoker/azure_openai_em_invoker.pyi

@@ -81,5 +81,5 @@ class AzureOpenAIEMInvoker(OpenAIEMInvoker):
             default_hyperparameters (dict[str, Any] | None, optional): Default hyperparameters for invoking the model.
                 Defaults to None.
             retry_config (RetryConfig | None, optional): The retry configuration for the embedding model.
-                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout is used.
+                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout will be used.
         """

gllm_inference/em_invoker/em_invoker.pyi

@@ -2,14 +2,11 @@ import abc
 from _typeshed import Incomplete
 from abc import ABC
 from gllm_core.utils.retry import RetryConfig
-from gllm_inference.constants import ALL_EXTENSIONS as ALL_EXTENSIONS, DOCUMENT_MIME_TYPES as DOCUMENT_MIME_TYPES
+from gllm_inference.constants import DOCUMENT_MIME_TYPES as DOCUMENT_MIME_TYPES, INVOKER_DEFAULT_TIMEOUT as INVOKER_DEFAULT_TIMEOUT
 from gllm_inference.exceptions import parse_error_message as parse_error_message
 from gllm_inference.schema import Attachment as Attachment, AttachmentType as AttachmentType, EMContent as EMContent, ModelId as ModelId, Vector as Vector
-from langchain_core.embeddings import Embeddings as Embeddings
 from typing import Any
 
-DEPRECATED_DETAILS: str
-
 class BaseEMInvoker(ABC, metaclass=abc.ABCMeta):
     """A base class for embedding model invokers used in Gen AI applications.
 
@@ -24,22 +21,18 @@ class BaseEMInvoker(ABC, metaclass=abc.ABCMeta):
     """
     default_hyperparameters: Incomplete
     retry_config: Incomplete
-    def __init__(self, model_id: ModelId, default_hyperparameters: dict[str, Any] | None = None, valid_extensions_map: dict[str, set[str]] | None = None, retry_config: RetryConfig | None = None, langchain_kwargs: dict[str, Any] | None = None) -> None:
-        '''Initializes a new instance of the BaseEMInvoker class.
+    def __init__(self, model_id: ModelId, default_hyperparameters: dict[str, Any] | None = None, supported_attachments: set[str] | None = None, retry_config: RetryConfig | None = None) -> None:
+        """Initializes a new instance of the BaseEMInvoker class.
 
         Args:
             model_id (ModelId): The model ID of the embedding model.
             default_hyperparameters (dict[str, Any] | None, optional): Default hyperparameters for invoking the
                 embedding model. Defaults to None, in which case an empty dictionary is used.
-            valid_extensions_map (dict[str, set[str]] | None, optional): A dictionary mapping for validating the
-                content type of the multimodal inputs. They keys are the mime types (e.g. "image") and the values are
-                the set of valid file extensions for the corresponding mime type. Defaults to None, in which case an
-                empty dictionary is used.
+            supported_attachments (set[str] | None, optional): A set of supported attachment types. Defaults to None,
+                in which case an empty set is used (indicating that no attachments are supported).
             retry_config (RetryConfig | None, optional): The retry configuration for the embedding model.
-                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout is used.
-            langchain_kwargs (dict[str, Any] | None, optional): Additional keyword arguments to initiate the LangChain
-                embedding model. Defaults to None.
-        '''
+                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout will be used.
+        """
     @property
     def model_id(self) -> str:
         """The model ID of the embedding model.
@@ -88,15 +81,3 @@ class BaseEMInvoker(ABC, metaclass=abc.ABCMeta):
             TimeoutError: If the invocation times out.
             ValueError: If the input content is invalid.
         """
-    def to_langchain(self) -> Embeddings:
-        """Converts the current embedding model invoker to an instance of LangChain `Embeddings` object.
-
-        This method converts the EM invoker to an instance of LangChain's `Embeddings` object.
-        This method requires the appropriate `langchain-<provider>` package to be installed.
-
-        Returns:
-            Embeddings: An instance of LangChain `Embeddings` object.
-
-        Raises:
-            ValueError: If `langchain_module_name` or `langchain_class_name` is missing.
-        """

gllm_inference/em_invoker/google_em_invoker.pyi

@@ -5,7 +5,7 @@ from gllm_inference.em_invoker.em_invoker import BaseEMInvoker as BaseEMInvoker
 from gllm_inference.schema import ModelId as ModelId, ModelProvider as ModelProvider, Vector as Vector
 from typing import Any
 
-VALID_EXTENSIONS_MAP: Incomplete
+SUPPORTED_ATTACHMENTS: Incomplete
 
 class GoogleEMInvoker(BaseEMInvoker):
     '''An embedding model invoker to interact with Google embedding models.
@@ -115,7 +115,7 @@ class GoogleEMInvoker(BaseEMInvoker):
             default_hyperparameters (dict[str, Any] | None, optional): Default hyperparameters for invoking the model.
                 Defaults to None.
             retry_config (RetryConfig | None, optional): The retry configuration for the embedding model.
-                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout is used.
+                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout will be used.
 
         Note:
             If neither `api_key` nor `credentials_path` is provided, Google Gen AI will be used by default.

gllm_inference/em_invoker/langchain/__init__.pyi

@@ -1,4 +1,3 @@
 from gllm_inference.em_invoker.langchain.em_invoker_embeddings import EMInvokerEmbeddings as EMInvokerEmbeddings
-from gllm_inference.em_invoker.langchain.tei_embeddings import TEIEmbeddings as TEIEmbeddings
 
-__all__ = ['EMInvokerEmbeddings', 'TEIEmbeddings']
+__all__ = ['EMInvokerEmbeddings']

gllm_inference/em_invoker/langchain_em_invoker.pyi

@@ -6,7 +6,7 @@ from gllm_inference.utils import load_langchain_model as load_langchain_model, p
 from langchain_core.embeddings import Embeddings as Embeddings
 from typing import Any
 
-VALID_EXTENSIONS_MAP: Incomplete
+SUPPORTED_ATTACHMENTS: Incomplete
 
 class LangChainEMInvoker(BaseEMInvoker):
     """A language model invoker to interact with LangChain's Embeddings.
@@ -19,7 +19,7 @@ class LangChainEMInvoker(BaseEMInvoker):
         retry_config (RetryConfig): The retry configuration for the embedding model.
     """
     model: Incomplete
-    def __init__(self, model: Embeddings | None = None, model_class_path: str | None = None, model_name: str | None = None, model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, retry_config: RetryConfig | None = None, em: Embeddings | None = None) -> None:
+    def __init__(self, model: Embeddings | None = None, model_class_path: str | None = None, model_name: str | None = None, model_kwargs: dict[str, Any] | None = None, default_hyperparameters: dict[str, Any] | None = None, retry_config: RetryConfig | None = None) -> None:
         '''Initializes a new instance of the LangChainEMInvoker class.
 
         Args:
@@ -35,15 +35,5 @@ class LangChainEMInvoker(BaseEMInvoker):
             default_hyperparameters (dict[str, Any] | None, optional): Default hyperparameters for invoking the model.
                 Defaults to None.
             retry_config (RetryConfig | None, optional): The retry configuration for the embedding model.
-                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout is used.
-            em (Embeddings | None, optional): Deprecated parameter to pass the LangChain\'s Embeddings instance.
-                Equivalent to the `model` parameter. Retained for backward compatibility. Defaults to None.
+                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout will be used.
         '''
-    def to_langchain(self) -> Embeddings:
-        """Converts the current embedding model invoker to an instance of LangChain `Embeddings` object.
-
-        This method converts the EM invoker to an instance of LangChain's `Embeddings` object.
-
-        Returns:
-            Embeddings: An instance of LangChain `Embeddings` object.
-        """

gllm_inference/em_invoker/openai_compatible_em_invoker.pyi

@@ -86,5 +86,5 @@ class OpenAICompatibleEMInvoker(OpenAIEMInvoker):
             default_hyperparameters (dict[str, Any] | None, optional): Default hyperparameters for invoking the model.
                 Defaults to None.
             retry_config (RetryConfig | None, optional): The retry configuration for the embedding model.
-                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout is used.
+                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout will be used.
         """

gllm_inference/em_invoker/openai_em_invoker.pyi

@@ -4,7 +4,7 @@ from gllm_inference.em_invoker.em_invoker import BaseEMInvoker as BaseEMInvoker
 from gllm_inference.schema import ModelId as ModelId, ModelProvider as ModelProvider, Vector as Vector
 from typing import Any
 
-VALID_EXTENSIONS_MAP: Incomplete
+SUPPORTED_ATTACHMENTS: Incomplete
 
 class OpenAIEMInvoker(BaseEMInvoker):
     '''An embedding model invoker to interact with OpenAI embedding models.
@@ -80,5 +80,5 @@ class OpenAIEMInvoker(BaseEMInvoker):
             default_hyperparameters (dict[str, Any] | None, optional): Default hyperparameters for invoking the model.
                 Defaults to None.
             retry_config (RetryConfig | None, optional): The retry configuration for the embedding model.
-                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout is used.
+                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout will be used.
         """

gllm_inference/em_invoker/twelevelabs_em_invoker.pyi

@@ -3,10 +3,9 @@ from gllm_core.utils.retry import RetryConfig as RetryConfig
 from gllm_inference.em_invoker.em_invoker import BaseEMInvoker as BaseEMInvoker
 from gllm_inference.em_invoker.schema.twelvelabs import InputType as InputType, Key as Key, OutputType as OutputType
 from gllm_inference.schema import Attachment as Attachment, AttachmentType as AttachmentType, EMContent as EMContent, ModelId as ModelId, ModelProvider as ModelProvider, Vector as Vector
-from langchain_core.embeddings import Embeddings as Embeddings
 from typing import Any
 
-VALID_EXTENSIONS_MAP: Incomplete
+SUPPORTED_ATTACHMENTS: Incomplete
 
 class TwelveLabsEMInvoker(BaseEMInvoker):
     '''An embedding model invoker to interact with TwelveLabs embedding models.
@@ -20,10 +19,7 @@ class TwelveLabsEMInvoker(BaseEMInvoker):
         retry_config (RetryConfig): The retry configuration for the embedding model.
 
     Input types:
-        The `TwelveLabsEMInvoker` supports the following input types:
-        1. Text.
-        2. Audio: ".mp3", ".wav", and ".flac".
-        3. Image: ".png", ".jpeg", and ".jpg".
+        The `TwelveLabsEMInvoker` supports the following input types: text, audio, and image.
         Non-text inputs must be passed as a `Attachment` object.
 
     Output format:
@@ -98,17 +94,5 @@ class TwelveLabsEMInvoker(BaseEMInvoker):
             default_hyperparameters (dict[str, Any] | None, optional): Default hyperparameters for invoking the model.
                 Defaults to None.
             retry_config (RetryConfig | None, optional): The retry configuration for the embedding model.
-                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout is used.
-        """
-    def to_langchain(self) -> Embeddings:
-        """Converts the current embedding model invoker to an instance of LangChain `Embeddings` object.
-
-        This method converts the EM invoker to an instance of LangChain's `Embeddings` object.
-        However, the TwelveLabsEMInvoker is not supported by LangChain.
-
-        Returns:
-            Embeddings: An instance of LangChain `Embeddings` object.
-
-        Raises:
-            NotImplementedError: The TwelveLabsEMInvoker is not supported by LangChain.
+                Defaults to None, in which case a default config with no retry and 30.0 seconds timeout will be used.
         """