@huggingface/tasks 0.3.2 → 0.3.4

package/dist/index.d.ts

@@ -549,7 +549,7 @@ interface ModelData {
             base_model_name?: string;
             task_type?: string;
         };
-        tokenizer?: TokenizerConfig;
+        tokenizer_config?: TokenizerConfig;
     };
     /**
      * all the model tags
@@ -575,7 +575,7 @@ interface ModelData {
      */
     widgetData?: WidgetExample[] | undefined;
     /**
-     * Parameters that will be used by the widget when calling Inference Endpoints (serverless)
+     * Parameters that will be used by the widget when calling Inference API (serverless)
      * https://huggingface.co/docs/api-inference/detailed_parameters
      *
      * can be set in the model card metadata (under `inference/parameters`)
@@ -732,6 +732,13 @@ declare const MODEL_LIBRARIES_UI_ELEMENTS: {
             };
         };
     };
+    audiocraft: {
+        prettyLabel: string;
+        repoName: string;
+        repoUrl: string;
+        snippets: (model: ModelData) => string[];
+        filter: false;
+    };
     bertopic: {
         prettyLabel: string;
         repoName: string;
@@ -1051,8 +1058,8 @@ declare const MODEL_LIBRARIES_UI_ELEMENTS: {
     };
 };
 type ModelLibraryKey = keyof typeof MODEL_LIBRARIES_UI_ELEMENTS;
-declare const ALL_MODEL_LIBRARY_KEYS: ("sklearn" | "adapter-transformers" | "allennlp" | "asteroid" | "bertopic" | "diffusers" | "doctr" | "espnet" | "fairseq" | "fastai" | "fasttext" | "flair" | "keras" | "k2" | "mindspore" | "ml-agents" | "mlx" | "nemo" | "open_clip" | "paddlenlp" | "peft" | "pyannote-audio" | "pythae" | "sample-factory" | "sentence-transformers" | "setfit" | "spacy" | "span-marker" | "speechbrain" | "stable-baselines3" | "stanza" | "tensorflowtts" | "timm" | "transformers" | "transformers.js" | "unity-sentis")[];
-declare const ALL_DISPLAY_MODEL_LIBRARY_KEYS: ("sklearn" | "adapter-transformers" | "allennlp" | "asteroid" | "bertopic" | "diffusers" | "doctr" | "espnet" | "fairseq" | "fastai" | "fasttext" | "flair" | "keras" | "k2" | "mindspore" | "ml-agents" | "mlx" | "nemo" | "open_clip" | "paddlenlp" | "peft" | "pyannote-audio" | "pythae" | "sample-factory" | "sentence-transformers" | "setfit" | "spacy" | "span-marker" | "speechbrain" | "stable-baselines3" | "stanza" | "tensorflowtts" | "timm" | "transformers" | "transformers.js" | "unity-sentis")[];
+declare const ALL_MODEL_LIBRARY_KEYS: ("sklearn" | "adapter-transformers" | "allennlp" | "asteroid" | "audiocraft" | "bertopic" | "diffusers" | "doctr" | "espnet" | "fairseq" | "fastai" | "fasttext" | "flair" | "keras" | "k2" | "mindspore" | "ml-agents" | "mlx" | "nemo" | "open_clip" | "paddlenlp" | "peft" | "pyannote-audio" | "pythae" | "sample-factory" | "sentence-transformers" | "setfit" | "spacy" | "span-marker" | "speechbrain" | "stable-baselines3" | "stanza" | "tensorflowtts" | "timm" | "transformers" | "transformers.js" | "unity-sentis")[];
+declare const ALL_DISPLAY_MODEL_LIBRARY_KEYS: ("sklearn" | "adapter-transformers" | "allennlp" | "asteroid" | "audiocraft" | "bertopic" | "diffusers" | "doctr" | "espnet" | "fairseq" | "fastai" | "fasttext" | "flair" | "keras" | "k2" | "mindspore" | "ml-agents" | "mlx" | "nemo" | "open_clip" | "paddlenlp" | "peft" | "pyannote-audio" | "pythae" | "sample-factory" | "sentence-transformers" | "setfit" | "spacy" | "span-marker" | "speechbrain" | "stable-baselines3" | "stanza" | "tensorflowtts" | "timm" | "transformers" | "transformers.js" | "unity-sentis")[];
 
 /**
  * Mapping from library name (excluding Transformers) to its supported tasks.
@@ -1066,6 +1073,1987 @@ declare const LIBRARY_TASK_MAPPING_EXCLUDING_TRANSFORMERS: Partial<Record<ModelL
 type PerLanguageMapping = Map<WidgetType, string[] | WidgetExample[]>;
 declare const MAPPING_DEFAULT_WIDGET: Map<string, PerLanguageMapping>;
 
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Audio Classification inference
+ */
+interface AudioClassificationInput {
+    /**
+     * The input audio data
+     */
+    inputs: unknown;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: AudioClassificationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Audio Classification
+ */
+interface AudioClassificationParameters {
+    function_to_apply?: ClassificationOutputTransform$3;
+    /**
+     * When specified, limits the output to the top K most probable classes.
+     */
+    top_k?: number;
+    [property: string]: unknown;
+}
+/**
+ * The function to apply to the model outputs in order to retrieve the scores.
+ */
+type ClassificationOutputTransform$3 = "sigmoid" | "softmax" | "none";
+type AudioClassificationOutput = AudioClassificationOutputElement[];
+/**
+ * Outputs for Audio Classification inference
+ */
+interface AudioClassificationOutputElement {
+    /**
+     * The predicted class label.
+     */
+    label: string;
+    /**
+     * The corresponding probability.
+     */
+    score: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Automatic Speech Recognition inference
+ */
+interface AutomaticSpeechRecognitionInput {
+    /**
+     * The input audio data
+     */
+    inputs: unknown;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: AutomaticSpeechRecognitionParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Automatic Speech Recognition
+ */
+interface AutomaticSpeechRecognitionParameters {
+    /**
+     * Parametrization of the text generation process
+     */
+    generate?: GenerationParameters$2;
+    /**
+     * Whether to output corresponding timestamps with the generated text
+     */
+    return_timestamps?: boolean;
+    [property: string]: unknown;
+}
+/**
+ * Parametrization of the text generation process
+ *
+ * Ad-hoc parametrization of the text generation process
+ */
+interface GenerationParameters$2 {
+    /**
+     * Whether to use sampling instead of greedy decoding when generating new tokens.
+     */
+    do_sample?: boolean;
+    /**
+     * Controls the stopping condition for beam-based methods.
+     */
+    early_stopping?: EarlyStoppingUnion$2;
+    /**
+     * If set to float strictly between 0 and 1, only tokens with a conditional probability
+     * greater than epsilon_cutoff will be sampled. In the paper, suggested values range from
+     * 3e-4 to 9e-4, depending on the size of the model. See [Truncation Sampling as Language
+     * Model Desmoothing](https://hf.co/papers/2210.15191) for more details.
+     */
+    epsilon_cutoff?: number;
+    /**
+     * Eta sampling is a hybrid of locally typical sampling and epsilon sampling. If set to
+     * float strictly between 0 and 1, a token is only considered if it is greater than either
+     * eta_cutoff or sqrt(eta_cutoff) * exp(-entropy(softmax(next_token_logits))). The latter
+     * term is intuitively the expected next token probability, scaled by sqrt(eta_cutoff). In
+     * the paper, suggested values range from 3e-4 to 2e-3, depending on the size of the model.
+     * See [Truncation Sampling as Language Model Desmoothing](https://hf.co/papers/2210.15191)
+     * for more details.
+     */
+    eta_cutoff?: number;
+    /**
+     * The maximum length (in tokens) of the generated text, including the input.
+     */
+    max_length?: number;
+    /**
+     * The maximum number of tokens to generate. Takes precedence over maxLength.
+     */
+    max_new_tokens?: number;
+    /**
+     * The minimum length (in tokens) of the generated text, including the input.
+     */
+    min_length?: number;
+    /**
+     * The minimum number of tokens to generate. Takes precedence over maxLength.
+     */
+    min_new_tokens?: number;
+    /**
+     * Number of groups to divide num_beams into in order to ensure diversity among different
+     * groups of beams. See [this paper](https://hf.co/papers/1610.02424) for more details.
+     */
+    num_beam_groups?: number;
+    /**
+     * Number of beams to use for beam search.
+     */
+    num_beams?: number;
+    /**
+     * The value balances the model confidence and the degeneration penalty in contrastive
+     * search decoding.
+     */
+    penalty_alpha?: number;
+    /**
+     * The value used to modulate the next token probabilities.
+     */
+    temperature?: number;
+    /**
+     * The number of highest probability vocabulary tokens to keep for top-k-filtering.
+     */
+    top_k?: number;
+    /**
+     * If set to float < 1, only the smallest set of most probable tokens with probabilities
+     * that add up to top_p or higher are kept for generation.
+     */
+    top_p?: number;
+    /**
+     * Local typicality measures how similar the conditional probability of predicting a target
+     * token next is to the expected conditional probability of predicting a random token next,
+     * given the partial text already generated. If set to float < 1, the smallest set of the
+     * most locally typical tokens with probabilities that add up to typical_p or higher are
+     * kept for generation. See [this paper](https://hf.co/papers/2202.00666) for more details.
+     */
+    typical_p?: number;
+    /**
+     * Whether the model should use the past last key/values attentions to speed up decoding
+     */
+    use_cache?: boolean;
+    [property: string]: unknown;
+}
+/**
+ * Controls the stopping condition for beam-based methods.
+ */
+type EarlyStoppingUnion$2 = boolean | "never";
+/**
+ * Outputs of inference for the Automatic Speech Recognition task
+ */
+interface AutomaticSpeechRecognitionOutput {
+    /**
+     * When returnTimestamps is enabled, chunks contains a list of audio chunks identified by
+     * the model.
+     */
+    chunks?: AutomaticSpeechRecognitionOutputChunk[];
+    /**
+     * The recognized text.
+     */
+    text: string;
+    [property: string]: unknown;
+}
+interface AutomaticSpeechRecognitionOutputChunk {
+    /**
+     * A chunk of text identified by the model
+     */
+    text: string;
+    /**
+     * The start and end timestamps corresponding with the text
+     */
+    timestamps: number[];
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Document Question Answering inference
+ */
+interface DocumentQuestionAnsweringInput {
+    /**
+     * One (document, question) pair to answer
+     */
+    inputs: DocumentQuestionAnsweringInputData;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: DocumentQuestionAnsweringParameters;
+    [property: string]: unknown;
+}
+/**
+ * One (document, question) pair to answer
+ */
+interface DocumentQuestionAnsweringInputData {
+    /**
+     * The image on which the question is asked
+     */
+    image: unknown;
+    /**
+     * A question to ask of the document
+     */
+    question: string;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Document Question Answering
+ */
+interface DocumentQuestionAnsweringParameters {
+    /**
+     * If the words in the document are too long to fit with the question for the model, it will
+     * be split in several chunks with some overlap. This argument controls the size of that
+     * overlap.
+     */
+    doc_stride?: number;
+    /**
+     * Whether to accept impossible as an answer
+     */
+    handle_impossible_answer?: boolean;
+    /**
+     * Language to use while running OCR. Defaults to english.
+     */
+    lang?: string;
+    /**
+     * The maximum length of predicted answers (e.g., only answers with a shorter length are
+     * considered).
+     */
+    max_answer_len?: number;
+    /**
+     * The maximum length of the question after tokenization. It will be truncated if needed.
+     */
+    max_question_len?: number;
+    /**
+     * The maximum length of the total sentence (context + question) in tokens of each chunk
+     * passed to the model. The context will be split in several chunks (using doc_stride as
+     * overlap) if needed.
+     */
+    max_seq_len?: number;
+    /**
+     * The number of answers to return (will be chosen by order of likelihood). Can return less
+     * than top_k answers if there are not enough options available within the context.
+     */
+    top_k?: number;
+    /**
+     * A list of words and bounding boxes (normalized 0->1000). If provided, the inference will
+     * skip the OCR step and use the provided bounding boxes instead.
+     */
+    word_boxes?: WordBox[];
+    [property: string]: unknown;
+}
+type WordBox = number[] | string;
+type DocumentQuestionAnsweringOutput = DocumentQuestionAnsweringOutputElement[];
+/**
+ * Outputs of inference for the Document Question Answering task
+ */
+interface DocumentQuestionAnsweringOutputElement {
+    /**
+     * The answer to the question.
+     */
+    answer: string;
+    /**
+     * The end word index of the answer (in the OCR’d version of the input or provided word
+     * boxes).
+     */
+    end: number;
+    /**
+     * The probability associated to the answer.
+     */
+    score: number;
+    /**
+     * The start word index of the answer (in the OCR’d version of the input or provided word
+     * boxes).
+     */
+    start: number;
+    /**
+     * The index of each word/box pair that is in the answer
+     */
+    words: number[];
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+type FeatureExtractionOutput = unknown[];
+/**
+ * Inputs for Text Embedding inference
+ */
+interface FeatureExtractionInput {
+    /**
+     * The text to get the embeddings of
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: {
+        [key: string]: unknown;
+    };
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Fill Mask inference
+ */
+interface FillMaskInput {
+    /**
+     * The text with masked tokens
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: FillMaskParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Fill Mask
+ */
+interface FillMaskParameters {
+    /**
+     * When passed, the model will limit the scores to the passed targets instead of looking up
+     * in the whole vocabulary. If the provided targets are not in the model vocab, they will be
+     * tokenized and the first resulting token will be used (with a warning, and that might be
+     * slower).
+     */
+    targets?: string[];
+    /**
+     * When passed, overrides the number of predictions to return.
+     */
+    top_k?: number;
+    [property: string]: unknown;
+}
+type FillMaskOutput = FillMaskOutputElement[];
+/**
+ * Outputs of inference for the Fill Mask task
+ */
+interface FillMaskOutputElement {
+    /**
+     * The corresponding probability
+     */
+    score: number;
+    /**
+     * The corresponding input with the mask token prediction.
+     */
+    sequence: string;
+    /**
+     * The predicted token id (to replace the masked one).
+     */
+    token: number;
+    tokenStr: unknown;
+    /**
+     * The predicted token (to replace the masked one).
+     */
+    token_str?: string;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Image Classification inference
+ */
+interface ImageClassificationInput {
+    /**
+     * The input image data
+     */
+    inputs: unknown;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: ImageClassificationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Image Classification
+ */
+interface ImageClassificationParameters {
+    function_to_apply?: ClassificationOutputTransform$2;
+    /**
+     * When specified, limits the output to the top K most probable classes.
+     */
+    top_k?: number;
+    [property: string]: unknown;
+}
+/**
+ * The function to apply to the model outputs in order to retrieve the scores.
+ */
+type ClassificationOutputTransform$2 = "sigmoid" | "softmax" | "none";
+type ImageClassificationOutput = ImageClassificationOutputElement[];
+/**
+ * Outputs of inference for the Image Classification task
+ */
+interface ImageClassificationOutputElement {
+    /**
+     * The predicted class label.
+     */
+    label: string;
+    /**
+     * The corresponding probability.
+     */
+    score: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Image To Image inference
+ */
+interface ImageToImageInput {
+    /**
+     * The input image data
+     */
+    inputs: unknown;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: ImageToImageParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Image To Image
+ */
+interface ImageToImageParameters {
+    /**
+     * For diffusion models. A higher guidance scale value encourages the model to generate
+     * images closely linked to the text prompt at the expense of lower image quality.
+     */
+    guidance_scale?: number;
+    /**
+     * One or several prompt to guide what NOT to include in image generation.
+     */
+    negative_prompt?: string[];
+    /**
+     * For diffusion models. The number of denoising steps. More denoising steps usually lead to
+     * a higher quality image at the expense of slower inference.
+     */
+    num_inference_steps?: number;
+    /**
+     * The size in pixel of the output image
+     */
+    target_size?: TargetSize$1;
+    [property: string]: unknown;
+}
+/**
+ * The size in pixel of the output image
+ */
+interface TargetSize$1 {
+    height: number;
+    width: number;
+    [property: string]: unknown;
+}
+/**
+ * Outputs of inference for the Image To Image task
+ */
+interface ImageToImageOutput {
+    /**
+     * The output image
+     */
+    image?: unknown;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Image To Text inference
+ */
+interface ImageToTextInput {
+    /**
+     * The input image data
+     */
+    inputs: unknown;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: ImageToTextParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Image To Text
+ */
+interface ImageToTextParameters {
+    /**
+     * Parametrization of the text generation process
+     */
+    generate?: GenerationParameters$1;
+    /**
+     * The amount of maximum tokens to generate.
+     */
+    max_new_tokens?: number;
+    [property: string]: unknown;
+}
+/**
+ * Parametrization of the text generation process
+ *
+ * Ad-hoc parametrization of the text generation process
+ */
+interface GenerationParameters$1 {
+    /**
+     * Whether to use sampling instead of greedy decoding when generating new tokens.
+     */
+    do_sample?: boolean;
+    /**
+     * Controls the stopping condition for beam-based methods.
+     */
+    early_stopping?: EarlyStoppingUnion$1;
+    /**
+     * If set to float strictly between 0 and 1, only tokens with a conditional probability
+     * greater than epsilon_cutoff will be sampled. In the paper, suggested values range from
+     * 3e-4 to 9e-4, depending on the size of the model. See [Truncation Sampling as Language
+     * Model Desmoothing](https://hf.co/papers/2210.15191) for more details.
+     */
+    epsilon_cutoff?: number;
+    /**
+     * Eta sampling is a hybrid of locally typical sampling and epsilon sampling. If set to
+     * float strictly between 0 and 1, a token is only considered if it is greater than either
+     * eta_cutoff or sqrt(eta_cutoff) * exp(-entropy(softmax(next_token_logits))). The latter
+     * term is intuitively the expected next token probability, scaled by sqrt(eta_cutoff). In
+     * the paper, suggested values range from 3e-4 to 2e-3, depending on the size of the model.
+     * See [Truncation Sampling as Language Model Desmoothing](https://hf.co/papers/2210.15191)
+     * for more details.
+     */
+    eta_cutoff?: number;
+    /**
+     * The maximum length (in tokens) of the generated text, including the input.
+     */
+    max_length?: number;
+    /**
+     * The maximum number of tokens to generate. Takes precedence over maxLength.
+     */
+    max_new_tokens?: number;
+    /**
+     * The minimum length (in tokens) of the generated text, including the input.
+     */
+    min_length?: number;
+    /**
+     * The minimum number of tokens to generate. Takes precedence over maxLength.
+     */
+    min_new_tokens?: number;
+    /**
+     * Number of groups to divide num_beams into in order to ensure diversity among different
+     * groups of beams. See [this paper](https://hf.co/papers/1610.02424) for more details.
+     */
+    num_beam_groups?: number;
+    /**
+     * Number of beams to use for beam search.
+     */
+    num_beams?: number;
+    /**
+     * The value balances the model confidence and the degeneration penalty in contrastive
+     * search decoding.
+     */
+    penalty_alpha?: number;
+    /**
+     * The value used to modulate the next token probabilities.
+     */
+    temperature?: number;
+    /**
+     * The number of highest probability vocabulary tokens to keep for top-k-filtering.
+     */
+    top_k?: number;
+    /**
+     * If set to float < 1, only the smallest set of most probable tokens with probabilities
+     * that add up to top_p or higher are kept for generation.
+     */
+    top_p?: number;
+    /**
+     * Local typicality measures how similar the conditional probability of predicting a target
+     * token next is to the expected conditional probability of predicting a random token next,
+     * given the partial text already generated. If set to float < 1, the smallest set of the
+     * most locally typical tokens with probabilities that add up to typical_p or higher are
+     * kept for generation. See [this paper](https://hf.co/papers/2202.00666) for more details.
+     */
+    typical_p?: number;
+    /**
+     * Whether the model should use the past last key/values attentions to speed up decoding
+     */
+    use_cache?: boolean;
+    [property: string]: unknown;
+}
+/**
+ * Controls the stopping condition for beam-based methods.
+ */
+type EarlyStoppingUnion$1 = boolean | "never";
+/**
+ * Outputs of inference for the Image To Text task
+ */
+interface ImageToTextOutput {
+    generatedText: unknown;
+    /**
+     * The generated text.
+     */
+    generated_text?: string;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Image Segmentation inference
+ */
+interface ImageSegmentationInput {
+    /**
+     * The input image data
+     */
+    inputs: unknown;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: ImageSegmentationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Image Segmentation
+ */
+interface ImageSegmentationParameters {
+    /**
+     * Threshold to use when turning the predicted masks into binary values.
+     */
+    mask_threshold?: number;
+    /**
+     * Mask overlap threshold to eliminate small, disconnected segments.
+     */
+    overlap_mask_area_threshold?: number;
+    /**
+     * Segmentation task to be performed, depending on model capabilities.
+     */
+    subtask?: ImageSegmentationSubtask;
+    /**
+     * Probability threshold to filter out predicted masks.
+     */
+    threshold?: number;
+    [property: string]: unknown;
+}
+type ImageSegmentationSubtask = "instance" | "panoptic" | "semantic";
+type ImageSegmentationOutput = ImageSegmentationOutputElement[];
+/**
+ * Outputs of inference for the Image Segmentation task
+ *
+ * A predicted mask / segment
+ */
+interface ImageSegmentationOutputElement {
+    /**
+     * The label of the predicted segment
+     */
+    label: string;
+    /**
+     * The corresponding mask as a black-and-white image
+     */
+    mask: unknown;
+    /**
+     * The score or confidence degreee the model has
+     */
+    score?: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Object Detection inference
+ */
+interface ObjectDetectionInput {
+    /**
+     * The input image data
+     */
+    inputs: unknown;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: ObjectDetectionParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Object Detection
+ */
+interface ObjectDetectionParameters {
+    /**
+     * The probability necessary to make a prediction.
+     */
+    threshold?: number;
+    [property: string]: unknown;
+}
+/**
+ * The predicted bounding box. Coordinates are relative to the top left corner of the input
+ * image.
+ */
+interface BoundingBox$1 {
+    xmax: number;
+    xmin: number;
+    ymax: number;
+    ymin: number;
+    [property: string]: unknown;
+}
+type ObjectDetectionOutput = ObjectDetectionOutputElement[];
+/**
+ * Outputs of inference for the Object Detection task
+ */
+interface ObjectDetectionOutputElement {
+    /**
+     * The predicted bounding box. Coordinates are relative to the top left corner of the input
+     * image.
+     */
+    box: BoundingBox$1;
+    /**
+     * The predicted label for the bounding box
+     */
+    label: string;
+    /**
+     * The associated score / probability
+     */
+    score: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Depth Estimation inference
+ */
+interface DepthEstimationInput {
+    /**
+     * The input image data
+     */
+    inputs: unknown;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: {
+        [key: string]: unknown;
+    };
+    [property: string]: unknown;
+}
+/**
+ * Outputs of inference for the Depth Estimation task
+ */
+interface DepthEstimationOutput {
+    /**
+     * The predicted depth as an image
+     */
+    depth?: unknown;
+    /**
+     * The predicted depth as a tensor
+     */
+    predicted_depth?: unknown;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Question Answering inference
+ */
+interface QuestionAnsweringInput {
+    /**
+     * One (context, question) pair to answer
+     */
+    inputs: QuestionAnsweringInputData;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: QuestionAnsweringParameters;
+    [property: string]: unknown;
+}
+/**
+ * One (context, question) pair to answer
+ */
+interface QuestionAnsweringInputData {
+    /**
+     * The context to be used for answering the question
+     */
+    context: string;
+    /**
+     * The question to be answered
+     */
+    question: string;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Question Answering
+ */
+interface QuestionAnsweringParameters {
+    /**
+     * Attempts to align the answer to real words. Improves quality on space separated
+     * languages. Might hurt on non-space-separated languages (like Japanese or Chinese)
+     */
+    align_to_words?: boolean;
+    /**
+     * If the context is too long to fit with the question for the model, it will be split in
+     * several chunks with some overlap. This argument controls the size of that overlap.
+     */
+    doc_stride?: number;
+    /**
+     * Whether to accept impossible as an answer.
+     */
+    handle_impossible_answer?: boolean;
+    /**
+     * The maximum length of predicted answers (e.g., only answers with a shorter length are
+     * considered).
+     */
+    max_answer_len?: number;
+    /**
+     * The maximum length of the question after tokenization. It will be truncated if needed.
+     */
+    max_question_len?: number;
+    /**
+     * The maximum length of the total sentence (context + question) in tokens of each chunk
+     * passed to the model. The context will be split in several chunks (using docStride as
+     * overlap) if needed.
+     */
+    max_seq_len?: number;
+    /**
+     * The number of answers to return (will be chosen by order of likelihood). Note that we
+     * return less than topk answers if there are not enough options available within the
+     * context.
+     */
+    top_k?: number;
+    [property: string]: unknown;
+}
+type QuestionAnsweringOutput = QuestionAnsweringOutputElement[];
+/**
+ * Outputs of inference for the Question Answering task
+ */
+interface QuestionAnsweringOutputElement {
+    /**
+     * The answer to the question.
+     */
+    answer: string;
+    /**
+     * The character position in the input where the answer ends.
+     */
+    end: number;
+    /**
+     * The probability associated to the answer.
+     */
+    score: number;
+    /**
+     * The character position in the input where the answer begins.
+     */
+    start: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+type SentenceSimilarityOutput = number[];
+/**
+ * Inputs for Sentence similarity inference
+ */
+interface SentenceSimilarityInput {
+    inputs: SentenceSimilarityInputData;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: {
+        [key: string]: unknown;
+    };
+    [property: string]: unknown;
+}
+interface SentenceSimilarityInputData {
+    /**
+     * A list of strings which will be compared against the source_sentence.
+     */
+    sentences: string[];
+    /**
+     * The string that you wish to compare the other strings with. This can be a phrase,
+     * sentence, or longer passage, depending on the model being used.
+     */
+    sourceSentence: string;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Summarization inference
+ *
+ * Inputs for Text2text Generation inference
+ */
+interface SummarizationInput {
+    /**
+     * The input text data
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: Text2TextGenerationParameters$1;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text2text Generation
+ */
+interface Text2TextGenerationParameters$1 {
+    /**
+     * Whether to clean up the potential extra spaces in the text output.
+     */
+    clean_up_tokenization_spaces?: boolean;
+    /**
+     * Additional parametrization of the text generation algorithm
+     */
+    generate_parameters?: {
+        [key: string]: unknown;
+    };
+    /**
+     * The truncation strategy to use
+     */
+    truncation?: Text2TextGenerationTruncationStrategy$1;
+    [property: string]: unknown;
+}
+type Text2TextGenerationTruncationStrategy$1 = "do_not_truncate" | "longest_first" | "only_first" | "only_second";
+/**
+ * Outputs of inference for the Summarization task
+ */
+interface SummarizationOutput {
+    /**
+     * The summarized text.
+     */
+    summary_text: string;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Table Question Answering inference
+ */
+interface TableQuestionAnsweringInput {
+    /**
+     * One (table, question) pair to answer
+     */
+    inputs: TableQuestionAnsweringInputData;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: {
+        [key: string]: unknown;
+    };
+    [property: string]: unknown;
+}
+/**
+ * One (table, question) pair to answer
+ */
+interface TableQuestionAnsweringInputData {
+    /**
+     * The question to be answered about the table
+     */
+    question: string;
+    /**
+     * The table to serve as context for the questions
+     */
+    table: {
+        [key: string]: string[];
+    };
+    [property: string]: unknown;
+}
+type TableQuestionAnsweringOutput = TableQuestionAnsweringOutputElement[];
+/**
+ * Outputs of inference for the Table Question Answering task
+ */
+interface TableQuestionAnsweringOutputElement {
+    /**
+     * If the model has an aggregator, this returns the aggregator.
+     */
+    aggregator?: string;
+    /**
+     * The answer of the question given the table. If there is an aggregator, the answer will be
+     * preceded by `AGGREGATOR >`.
+     */
+    answer: string;
+    /**
+     * List of strings made up of the answer cell values.
+     */
+    cells: string[];
+    /**
+     * Coordinates of the cells of the answers.
+     */
+    coordinates: Array<number[]>;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Text To Image inference
+ */
+interface TextToImageInput {
+    /**
+     * The input text data (sometimes called "prompt"
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TextToImageParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text To Image
+ */
+interface TextToImageParameters {
+    /**
+     * For diffusion models. A higher guidance scale value encourages the model to generate
+     * images closely linked to the text prompt at the expense of lower image quality.
+     */
+    guidance_scale?: number;
+    /**
+     * One or several prompt to guide what NOT to include in image generation.
+     */
+    negative_prompt?: string[];
+    /**
+     * For diffusion models. The number of denoising steps. More denoising steps usually lead to
+     * a higher quality image at the expense of slower inference.
+     */
+    num_inference_steps?: number;
+    /**
+     * For diffusion models. Override the scheduler with a compatible one
+     */
+    scheduler?: string;
+    /**
+     * The size in pixel of the output image
+     */
+    target_size?: TargetSize;
+    [property: string]: unknown;
+}
+/**
+ * The size in pixel of the output image
+ */
+interface TargetSize {
+    height: number;
+    width: number;
+    [property: string]: unknown;
+}
+/**
+ * Outputs of inference for the Text To Image task
+ */
+interface TextToImageOutput {
+    /**
+     * The generated image
+     */
+    image: unknown;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Text to Speech inference
+ *
+ * Inputs for Text To Audio inference
+ */
+interface TextToSpeechInput {
+    /**
+     * The input text data
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TextToAudioParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text To Audio
+ */
+interface TextToAudioParameters {
+    /**
+     * Parametrization of the text generation process
+     */
+    generate?: GenerationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Parametrization of the text generation process
+ *
+ * Ad-hoc parametrization of the text generation process
+ */
+interface GenerationParameters {
+    /**
+     * Whether to use sampling instead of greedy decoding when generating new tokens.
+     */
+    do_sample?: boolean;
+    /**
+     * Controls the stopping condition for beam-based methods.
+     */
+    early_stopping?: EarlyStoppingUnion;
+    /**
+     * If set to float strictly between 0 and 1, only tokens with a conditional probability
+     * greater than epsilon_cutoff will be sampled. In the paper, suggested values range from
+     * 3e-4 to 9e-4, depending on the size of the model. See [Truncation Sampling as Language
+     * Model Desmoothing](https://hf.co/papers/2210.15191) for more details.
+     */
+    epsilon_cutoff?: number;
+    /**
+     * Eta sampling is a hybrid of locally typical sampling and epsilon sampling. If set to
+     * float strictly between 0 and 1, a token is only considered if it is greater than either
+     * eta_cutoff or sqrt(eta_cutoff) * exp(-entropy(softmax(next_token_logits))). The latter
+     * term is intuitively the expected next token probability, scaled by sqrt(eta_cutoff). In
+     * the paper, suggested values range from 3e-4 to 2e-3, depending on the size of the model.
+     * See [Truncation Sampling as Language Model Desmoothing](https://hf.co/papers/2210.15191)
+     * for more details.
+     */
+    eta_cutoff?: number;
+    /**
+     * The maximum length (in tokens) of the generated text, including the input.
+     */
+    max_length?: number;
+    /**
+     * The maximum number of tokens to generate. Takes precedence over maxLength.
+     */
+    max_new_tokens?: number;
+    /**
+     * The minimum length (in tokens) of the generated text, including the input.
+     */
+    min_length?: number;
+    /**
+     * The minimum number of tokens to generate. Takes precedence over maxLength.
+     */
+    min_new_tokens?: number;
+    /**
+     * Number of groups to divide num_beams into in order to ensure diversity among different
+     * groups of beams. See [this paper](https://hf.co/papers/1610.02424) for more details.
+     */
+    num_beam_groups?: number;
+    /**
+     * Number of beams to use for beam search.
+     */
+    num_beams?: number;
+    /**
+     * The value balances the model confidence and the degeneration penalty in contrastive
+     * search decoding.
+     */
+    penalty_alpha?: number;
+    /**
+     * The value used to modulate the next token probabilities.
+     */
+    temperature?: number;
+    /**
+     * The number of highest probability vocabulary tokens to keep for top-k-filtering.
+     */
+    top_k?: number;
+    /**
+     * If set to float < 1, only the smallest set of most probable tokens with probabilities
+     * that add up to top_p or higher are kept for generation.
+     */
+    top_p?: number;
+    /**
+     * Local typicality measures how similar the conditional probability of predicting a target
+     * token next is to the expected conditional probability of predicting a random token next,
+     * given the partial text already generated. If set to float < 1, the smallest set of the
+     * most locally typical tokens with probabilities that add up to typical_p or higher are
+     * kept for generation. See [this paper](https://hf.co/papers/2202.00666) for more details.
+     */
+    typical_p?: number;
+    /**
+     * Whether the model should use the past last key/values attentions to speed up decoding
+     */
+    use_cache?: boolean;
+    [property: string]: unknown;
+}
+/**
+ * Controls the stopping condition for beam-based methods.
+ */
+type EarlyStoppingUnion = boolean | "never";
+/**
+ * Outputs for Text to Speech inference
+ *
+ * Outputs of inference for the Text To Audio task
+ */
+interface TextToSpeechOutput {
+    /**
+     * The generated audio waveform.
+     */
+    audio: unknown;
+    samplingRate: unknown;
+    /**
+     * The sampling rate of the generated audio waveform.
+     */
+    sampling_rate?: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Token Classification inference
+ */
+interface TokenClassificationInput {
+    /**
+     * The input text data
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TokenClassificationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Token Classification
+ */
+interface TokenClassificationParameters {
+    /**
+     * The strategy used to fuse tokens based on model predictions
+     */
+    aggregation_strategy?: TokenClassificationAggregationStrategy;
+    /**
+     * A list of labels to ignore
+     */
+    ignore_labels?: string[];
+    /**
+     * The number of overlapping tokens between chunks when splitting the input text.
+     */
+    stride?: number;
+    [property: string]: unknown;
+}
+/**
+ * Do not aggregate tokens
+ *
+ * Group consecutive tokens with the same label in a single entity.
+ *
+ * Similar to "simple", also preserves word integrity (use the label predicted for the first
+ * token in a word).
+ *
+ * Similar to "simple", also preserves word integrity (uses the label with the highest
+ * score, averaged across the word's tokens).
+ *
+ * Similar to "simple", also preserves word integrity (uses the label with the highest score
+ * across the word's tokens).
+ */
+type TokenClassificationAggregationStrategy = "none" | "simple" | "first" | "average" | "max";
+type TokenClassificationOutput = TokenClassificationOutputElement[];
+/**
+ * Outputs of inference for the Token Classification task
+ */
+interface TokenClassificationOutputElement {
+    /**
+     * The character position in the input where this group ends.
+     */
+    end?: number;
+    /**
+     * The predicted label for that group of tokens
+     */
+    entity_group?: string;
+    label: unknown;
+    /**
+     * The associated score / probability
+     */
+    score: number;
+    /**
+     * The character position in the input where this group begins.
+     */
+    start?: number;
+    /**
+     * The corresponding text
+     */
+    word?: string;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Translation inference
+ *
+ * Inputs for Text2text Generation inference
+ */
+interface TranslationInput {
+    /**
+     * The input text data
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: Text2TextGenerationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text2text Generation
+ */
+interface Text2TextGenerationParameters {
+    /**
+     * Whether to clean up the potential extra spaces in the text output.
+     */
+    clean_up_tokenization_spaces?: boolean;
+    /**
+     * Additional parametrization of the text generation algorithm
+     */
+    generate_parameters?: {
+        [key: string]: unknown;
+    };
+    /**
+     * The truncation strategy to use
+     */
+    truncation?: Text2TextGenerationTruncationStrategy;
+    [property: string]: unknown;
+}
+type Text2TextGenerationTruncationStrategy = "do_not_truncate" | "longest_first" | "only_first" | "only_second";
+/**
+ * Outputs of inference for the Translation task
+ */
+interface TranslationOutput {
+    /**
+     * The translated text.
+     */
+    translation_text: string;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Text Classification inference
+ */
+interface TextClassificationInput {
+    /**
+     * The text to classify
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TextClassificationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text Classification
+ */
+interface TextClassificationParameters {
+    function_to_apply?: ClassificationOutputTransform$1;
+    /**
+     * When specified, limits the output to the top K most probable classes.
+     */
+    top_k?: number;
+    [property: string]: unknown;
+}
+/**
+ * The function to apply to the model outputs in order to retrieve the scores.
+ */
+type ClassificationOutputTransform$1 = "sigmoid" | "softmax" | "none";
+type TextClassificationOutput = TextClassificationOutputElement[];
+/**
+ * Outputs of inference for the Text Classification task
+ */
+interface TextClassificationOutputElement {
+    /**
+     * The predicted class label.
+     */
+    label: string;
+    /**
+     * The corresponding probability.
+     */
+    score: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Text Generation inference
+ */
+interface TextGenerationInput {
+    /**
+     * The text to initialize generation with
+     */
+    inputs: string;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: TextGenerationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Text Generation
+ */
+interface TextGenerationParameters {
+    /**
+     * The number of sampling queries to run. Only the best one (in terms of total logprob) will
+     * be returned.
+     */
+    best_of?: number;
+    /**
+     * Whether or not to output decoder input details
+     */
+    decoder_input_details?: boolean;
+    /**
+     * Whether or not to output details
+     */
+    details?: boolean;
+    /**
+     * Whether to use logits sampling instead of greedy decoding when generating new tokens.
+     */
+    do_sample?: boolean;
+    /**
+     * The maximum number of tokens to generate.
+     */
+    max_new_tokens?: number;
+    /**
+     * The parameter for repetition penalty. A value of 1.0 means no penalty. See [this
+     * paper](https://hf.co/papers/1909.05858) for more details.
+     */
+    repetition_penalty?: number;
+    /**
+     * Whether to prepend the prompt to the generated text.
+     */
+    return_full_text?: boolean;
+    /**
+     * The random sampling seed.
+     */
+    seed?: number;
+    /**
+     * Stop generating tokens if a member of `stop_sequences` is generated.
+     */
+    stop_sequences?: string[];
+    /**
+     * The value used to modulate the logits distribution.
+     */
+    temperature?: number;
+    /**
+     * The number of highest probability vocabulary tokens to keep for top-k-filtering.
+     */
+    top_k?: number;
+    /**
+     * If set to < 1, only the smallest set of most probable tokens with probabilities that add
+     * up to `top_p` or higher are kept for generation.
+     */
+    top_p?: number;
+    /**
+     * Truncate input tokens to the given size.
+     */
+    truncate?: number;
+    /**
+     * Typical Decoding mass. See [Typical Decoding for Natural Language
+     * Generation](https://hf.co/papers/2202.00666) for more information
+     */
+    typical_p?: number;
+    /**
+     * Watermarking with [A Watermark for Large Language Models](https://hf.co/papers/2301.10226)
+     */
+    watermark?: boolean;
+    [property: string]: unknown;
+}
+/**
+ * Outputs for Text Generation inference
+ */
+interface TextGenerationOutput {
+    /**
+     * When enabled, details about the generation
+     */
+    details?: TextGenerationOutputDetails;
+    /**
+     * The generated text
+     */
+    generated_text: string;
+    [property: string]: unknown;
+}
+/**
+ * When enabled, details about the generation
+ */
+interface TextGenerationOutputDetails {
+    /**
+     * Details about additional sequences when best_of is provided
+     */
+    best_of_sequences?: TextGenerationSequenceDetails[];
+    /**
+     * The reason why the generation was stopped.
+     */
+    finish_reason: FinishReason;
+    /**
+     * The number of generated tokens
+     */
+    generated_tokens: number;
+    prefill: PrefillToken[];
+    /**
+     * The random seed used for generation
+     */
+    seed?: number;
+    /**
+     * The generated tokens and associated details
+     */
+    tokens: Token[];
+    [property: string]: unknown;
+}
+interface TextGenerationSequenceDetails {
+    /**
+     * The reason why the generation was stopped.
+     */
+    finish_reason: FinishReason;
+    /**
+     * The generated text
+     */
+    generated_text: number;
+    /**
+     * The number of generated tokens
+     */
+    generated_tokens: number;
+    prefill: PrefillToken[];
+    /**
+     * The random seed used for generation
+     */
+    seed?: number;
+    /**
+     * The generated tokens and associated details
+     */
+    tokens: Token[];
+    [property: string]: unknown;
+}
+/**
+ * The generated sequence reached the maximum allowed length
+ *
+ * The model generated an end-of-sentence (EOS) token
+ *
+ * One of the sequence in stop_sequences was generated
+ */
+type FinishReason = "length" | "eos_token" | "stop_sequence";
+interface PrefillToken {
+    id: number;
+    logprob: number;
+    /**
+     * The text associated with that token
+     */
+    text: string;
+    [property: string]: unknown;
+}
+interface Token {
+    id: number;
+    logprob: number;
+    /**
+     * Whether or not that token is a special one
+     */
+    special: boolean;
+    /**
+     * The text associated with that token
+     */
+    text: string;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Video Classification inference
+ */
+interface VideoClassificationInput {
+    /**
+     * The input video data
+     */
+    inputs: unknown;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: VideoClassificationParameters;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Video Classification
+ */
+interface VideoClassificationParameters {
+    /**
+     * The sampling rate used to select frames from the video.
+     */
+    frame_sampling_rate?: number;
+    function_to_apply?: ClassificationOutputTransform;
+    /**
+     * The number of sampled frames to consider for classification.
+     */
+    num_frames?: number;
+    /**
+     * When specified, limits the output to the top K most probable classes.
+     */
+    top_k?: number;
+    [property: string]: unknown;
+}
+/**
+ * The function to apply to the model outputs in order to retrieve the scores.
+ */
+type ClassificationOutputTransform = "sigmoid" | "softmax" | "none";
+type VideoClassificationOutput = VideoClassificationOutputElement[];
+/**
+ * Outputs of inference for the Video Classification task
+ */
+interface VideoClassificationOutputElement {
+    /**
+     * The predicted class label.
+     */
+    label: string;
+    /**
+     * The corresponding probability.
+     */
+    score: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Visual Question Answering inference
+ */
+interface VisualQuestionAnsweringInput {
+    /**
+     * One (image, question) pair to answer
+     */
+    inputs: VisualQuestionAnsweringInputData;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: VisualQuestionAnsweringParameters;
+    [property: string]: unknown;
+}
+/**
+ * One (image, question) pair to answer
+ */
+interface VisualQuestionAnsweringInputData {
+    /**
+     * The image.
+     */
+    image: unknown;
+    /**
+     * The question to answer based on the image.
+     */
+    question: unknown;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Visual Question Answering
+ */
+interface VisualQuestionAnsweringParameters {
+    /**
+     * The number of answers to return (will be chosen by order of likelihood). Note that we
+     * return less than topk answers if there are not enough options available within the
+     * context.
+     */
+    top_k?: number;
+    [property: string]: unknown;
+}
+type VisualQuestionAnsweringOutput = VisualQuestionAnsweringOutputElement[];
+/**
+ * Outputs of inference for the Visual Question Answering task
+ */
+interface VisualQuestionAnsweringOutputElement {
+    /**
+     * The answer to the question
+     */
+    answer?: string;
+    label: unknown;
+    /**
+     * The associated score / probability
+     */
+    score: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Zero Shot Classification inference
+ */
+interface ZeroShotClassificationInput {
+    /**
+     * The input text data, with candidate labels
+     */
+    inputs: ZeroShotClassificationInputData;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: ZeroShotClassificationParameters;
+    [property: string]: unknown;
+}
+/**
+ * The input text data, with candidate labels
+ */
+interface ZeroShotClassificationInputData {
+    /**
+     * The set of possible class labels to classify the text into.
+     */
+    candidateLabels: string[];
+    /**
+     * The text to classify
+     */
+    text: string;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Zero Shot Classification
+ */
+interface ZeroShotClassificationParameters {
+    /**
+     * The sentence used in conjunction with candidateLabels to attempt the text classification
+     * by replacing the placeholder with the candidate labels.
+     */
+    hypothesis_template?: string;
+    /**
+     * Whether multiple candidate labels can be true. If false, the scores are normalized such
+     * that the sum of the label likelihoods for each sequence is 1. If true, the labels are
+     * considered independent and probabilities are normalized for each candidate.
+     */
+    multi_label?: boolean;
+    [property: string]: unknown;
+}
+type ZeroShotClassificationOutput = ZeroShotClassificationOutputElement[];
+/**
+ * Outputs of inference for the Zero Shot Classification task
+ */
+interface ZeroShotClassificationOutputElement {
+    /**
+     * The predicted class label.
+     */
+    label: string;
+    /**
+     * The corresponding probability.
+     */
+    score: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Zero Shot Image Classification inference
+ */
+interface ZeroShotImageClassificationInput {
+    /**
+     * The input image data, with candidate labels
+     */
+    inputs: ZeroShotImageClassificationInputData;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: ZeroShotImageClassificationParameters;
+    [property: string]: unknown;
+}
+/**
+ * The input image data, with candidate labels
+ */
+interface ZeroShotImageClassificationInputData {
+    /**
+     * The candidate labels for this image
+     */
+    candidateLabels: string[];
+    /**
+     * The image data to classify
+     */
+    image: unknown;
+    [property: string]: unknown;
+}
+/**
+ * Additional inference parameters
+ *
+ * Additional inference parameters for Zero Shot Image Classification
+ */
+interface ZeroShotImageClassificationParameters {
+    /**
+     * The sentence used in conjunction with candidateLabels to attempt the text classification
+     * by replacing the placeholder with the candidate labels.
+     */
+    hypothesis_template?: string;
+    [property: string]: unknown;
+}
+type ZeroShotImageClassificationOutput = ZeroShotImageClassificationOutputElement[];
+/**
+ * Outputs of inference for the Zero Shot Image Classification task
+ */
+interface ZeroShotImageClassificationOutputElement {
+    /**
+     * The predicted class label.
+     */
+    label: string;
+    /**
+     * The corresponding probability.
+     */
+    score: number;
+    [property: string]: unknown;
+}
+
+/**
+ * Inference code generated from the JSON schema spec in ./spec
+ *
+ * Using src/scripts/inference-codegen
+ */
+/**
+ * Inputs for Zero Shot Object Detection inference
+ */
+interface ZeroShotObjectDetectionInput {
+    /**
+     * The input image data, with candidate labels
+     */
+    inputs: ZeroShotObjectDetectionInputData;
+    /**
+     * Additional inference parameters
+     */
+    parameters?: {
+        [key: string]: unknown;
+    };
+    [property: string]: unknown;
+}
+/**
+ * The input image data, with candidate labels
+ */
+interface ZeroShotObjectDetectionInputData {
+    /**
+     * The candidate labels for this image
+     */
+    candidateLabels: string[];
+    /**
+     * The image data to generate bounding boxes from
+     */
+    image: unknown;
+    [property: string]: unknown;
+}
+/**
+ * The predicted bounding box. Coordinates are relative to the top left corner of the input
+ * image.
+ */
+interface BoundingBox {
+    xmax: number;
+    xmin: number;
+    ymax: number;
+    ymin: number;
+    [property: string]: unknown;
+}
+type ZeroShotObjectDetectionOutput = ZeroShotObjectDetectionOutputElement[];
+/**
+ * Outputs of inference for the Zero Shot Object Detection task
+ */
+interface ZeroShotObjectDetectionOutputElement {
+    /**
+     * The predicted bounding box. Coordinates are relative to the top left corner of the input
+     * image.
+     */
+    box: BoundingBox;
+    /**
+     * A candidate label
+     */
+    label: string;
+    /**
+     * The associated score / probability
+     */
+    score: number;
+    [property: string]: unknown;
+}
+
 /**
  * Model libraries compatible with each ML task
  */
@@ -1231,4 +3219,4 @@ declare namespace index {
   };
 }
 
-export { ALL_DISPLAY_MODEL_LIBRARY_KEYS, ALL_MODEL_LIBRARY_KEYS, ExampleRepo, InferenceDisplayability, LIBRARY_TASK_MAPPING_EXCLUDING_TRANSFORMERS, LibraryUiElement, MAPPING_DEFAULT_WIDGET, MODALITIES, MODALITY_LABELS, MODEL_LIBRARIES_UI_ELEMENTS, Modality, ModelData, ModelLibraryKey, PIPELINE_DATA, PIPELINE_TYPES, PIPELINE_TYPES_SET, PipelineData, PipelineType, SPECIAL_TOKENS_ATTRIBUTES, SUBTASK_TYPES, SpecialTokensMap, TASKS_DATA, TASKS_MODEL_LIBRARIES, TaskData, TaskDataCustom, TaskDemo, TaskDemoEntry, TokenizerConfig, TransformersInfo, WidgetExample, WidgetExampleAssetAndPromptInput, WidgetExampleAssetAndTextInput, WidgetExampleAssetAndZeroShotInput, WidgetExampleAssetInput, WidgetExampleAttribute, WidgetExampleOutput, WidgetExampleOutputAnswerScore, WidgetExampleOutputLabels, WidgetExampleOutputText, WidgetExampleOutputUrl, WidgetExampleSentenceSimilarityInput, WidgetExampleStructuredDataInput, WidgetExampleTableDataInput, WidgetExampleTextAndContextInput, WidgetExampleTextAndTableInput, WidgetExampleTextInput, WidgetExampleZeroShotTextInput, WidgetType, index as snippets };
+export { ALL_DISPLAY_MODEL_LIBRARY_KEYS, ALL_MODEL_LIBRARY_KEYS, AudioClassificationInput, AudioClassificationOutput, AudioClassificationOutputElement, AudioClassificationParameters, AutomaticSpeechRecognitionInput, AutomaticSpeechRecognitionOutput, AutomaticSpeechRecognitionOutputChunk, AutomaticSpeechRecognitionParameters, BoundingBox, ClassificationOutputTransform$1 as ClassificationOutputTransform, DepthEstimationInput, DepthEstimationOutput, DocumentQuestionAnsweringInput, DocumentQuestionAnsweringInputData, DocumentQuestionAnsweringOutput, DocumentQuestionAnsweringOutputElement, DocumentQuestionAnsweringParameters, EarlyStoppingUnion$2 as EarlyStoppingUnion, ExampleRepo, FeatureExtractionInput, FeatureExtractionOutput, FillMaskInput, FillMaskOutput, FillMaskOutputElement, FillMaskParameters, FinishReason, GenerationParameters$2 as GenerationParameters, ImageClassificationInput, ImageClassificationOutput, ImageClassificationOutputElement, ImageClassificationParameters, ImageSegmentationInput, ImageSegmentationOutput, ImageSegmentationOutputElement, ImageSegmentationParameters, ImageSegmentationSubtask, ImageToImageInput, ImageToImageOutput, ImageToImageParameters, ImageToTextInput, ImageToTextOutput, ImageToTextParameters, InferenceDisplayability, LIBRARY_TASK_MAPPING_EXCLUDING_TRANSFORMERS, LibraryUiElement, MAPPING_DEFAULT_WIDGET, MODALITIES, MODALITY_LABELS, MODEL_LIBRARIES_UI_ELEMENTS, Modality, ModelData, ModelLibraryKey, ObjectDetectionInput, ObjectDetectionOutput, ObjectDetectionOutputElement, ObjectDetectionParameters, PIPELINE_DATA, PIPELINE_TYPES, PIPELINE_TYPES_SET, PipelineData, PipelineType, PrefillToken, QuestionAnsweringInput, QuestionAnsweringInputData, QuestionAnsweringOutput, QuestionAnsweringOutputElement, QuestionAnsweringParameters, SPECIAL_TOKENS_ATTRIBUTES, SUBTASK_TYPES, SentenceSimilarityInput, SentenceSimilarityInputData, SentenceSimilarityOutput, SpecialTokensMap, SummarizationInput, SummarizationOutput, TASKS_DATA, TASKS_MODEL_LIBRARIES, TableQuestionAnsweringInput, TableQuestionAnsweringInputData, TableQuestionAnsweringOutput, TableQuestionAnsweringOutputElement, TargetSize$1 as TargetSize, TaskData, TaskDataCustom, TaskDemo, TaskDemoEntry, Text2TextGenerationParameters, Text2TextGenerationTruncationStrategy, TextClassificationInput, TextClassificationOutput, TextClassificationOutputElement, TextClassificationParameters, TextGenerationInput, TextGenerationOutput, TextGenerationOutputDetails, TextGenerationParameters, TextGenerationSequenceDetails, TextToAudioParameters, TextToImageInput, TextToImageOutput, TextToImageParameters, TextToSpeechInput, TextToSpeechOutput, Token, TokenClassificationAggregationStrategy, TokenClassificationInput, TokenClassificationOutput, TokenClassificationOutputElement, TokenClassificationParameters, TokenizerConfig, TransformersInfo, TranslationInput, TranslationOutput, VideoClassificationInput, VideoClassificationOutput, VideoClassificationOutputElement, VideoClassificationParameters, VisualQuestionAnsweringInput, VisualQuestionAnsweringInputData, VisualQuestionAnsweringOutput, VisualQuestionAnsweringOutputElement, VisualQuestionAnsweringParameters, WidgetExample, WidgetExampleAssetAndPromptInput, WidgetExampleAssetAndTextInput, WidgetExampleAssetAndZeroShotInput, WidgetExampleAssetInput, WidgetExampleAttribute, WidgetExampleOutput, WidgetExampleOutputAnswerScore, WidgetExampleOutputLabels, WidgetExampleOutputText, WidgetExampleOutputUrl, WidgetExampleSentenceSimilarityInput, WidgetExampleStructuredDataInput, WidgetExampleTableDataInput, WidgetExampleTextAndContextInput, WidgetExampleTextAndTableInput, WidgetExampleTextInput, WidgetExampleZeroShotTextInput, WidgetType, WordBox, ZeroShotClassificationInput, ZeroShotClassificationInputData, ZeroShotClassificationOutput, ZeroShotClassificationOutputElement, ZeroShotClassificationParameters, ZeroShotImageClassificationInput, ZeroShotImageClassificationInputData, ZeroShotImageClassificationOutput, ZeroShotImageClassificationOutputElement, ZeroShotImageClassificationParameters, ZeroShotObjectDetectionInput, ZeroShotObjectDetectionInputData, ZeroShotObjectDetectionOutput, ZeroShotObjectDetectionOutputElement, index as snippets };