Inference types
This page lists the types (e.g. dataclasses) available for each task supported on the Hugging Face Hub. Each task is specified using a JSON schema, and the types are generated from these schemas - with some customization due to Python requirements. Visit @huggingface.js/tasks to find the JSON schemas for each task.
This part of the lib is still under development and will be improved in future releases.
audio_classification
class huggingface_hub.AudioClassificationInput
< source >( inputs: Any parameters: Optional = None )
Inputs for Audio Classification inference
Outputs for Audio Classification inference
class huggingface_hub.AudioClassificationParameters
< source >( function_to_apply: Optional = None top_k: Optional = None )
Additional inference parameters Additional inference parameters for Audio Classification
audio_to_audio
Inputs for Audio to Audio inference
class huggingface_hub.AudioToAudioOutputElement
< source >( blob: Any content_type: str label: str )
Outputs of inference for the Audio To Audio task A generated audio file with its label.
automatic_speech_recognition
class huggingface_hub.AutomaticSpeechRecognitionGenerationParameters
< source >( do_sample: Optional = None early_stopping: Union = None epsilon_cutoff: Optional = None eta_cutoff: Optional = None max_length: Optional = None max_new_tokens: Optional = None min_length: Optional = None min_new_tokens: Optional = None num_beam_groups: Optional = None num_beams: Optional = None penalty_alpha: Optional = None temperature: Optional = None top_k: Optional = None top_p: Optional = None typical_p: Optional = None use_cache: Optional = None )
Parametrization of the text generation process Ad-hoc parametrization of the text generation process
class huggingface_hub.AutomaticSpeechRecognitionInput
< source >( inputs: Any parameters: Optional = None )
Inputs for Automatic Speech Recognition inference
class huggingface_hub.AutomaticSpeechRecognitionOutput
< source >( text: str chunks: Optional = None )
Outputs of inference for the Automatic Speech Recognition task
class huggingface_hub.AutomaticSpeechRecognitionOutputChunk
< source >( text: str timestamps: List )
class huggingface_hub.AutomaticSpeechRecognitionParameters
< source >( generate: Optional = None return_timestamps: Optional = None )
Additional inference parameters Additional inference parameters for Automatic Speech Recognition
chat_completion
class huggingface_hub.ChatCompletionInput
< source >( messages: List frequency_penalty: Optional = None logit_bias: Optional = None logprobs: Optional = None max_tokens: Optional = None model: Optional = None n: Optional = None presence_penalty: Optional = None response_format: Optional = None seed: Optional = None stop: Optional = None stream: Optional = None temperature: Optional = None tool_choice: Union = None tool_prompt: Optional = None tools: Optional = None top_logprobs: Optional = None top_p: Optional = None )
Chat Completion Input. Auto-generated from TGI specs. For more details, check out https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts.
class huggingface_hub.ChatCompletionInputFunctionDefinition
< source >( arguments: Any name: str description: Optional = None )
class huggingface_hub.ChatCompletionInputGrammarType
< source >( type: ChatCompletionInputGrammarTypeType value: Any )
class huggingface_hub.ChatCompletionInputMessage
< source >( content: Union role: str name: Optional = None )
class huggingface_hub.ChatCompletionInputMessageChunk
< source >( type: ChatCompletionInputMessageChunkType image_url: Optional = None text: Optional = None )
class huggingface_hub.ChatCompletionInputTool
< source >( function: ChatCompletionInputFunctionDefinition type: str )
class huggingface_hub.ChatCompletionOutput
< source >( choices: List created: int id: str model: str system_fingerprint: str usage: ChatCompletionOutputUsage )
Chat Completion Output. Auto-generated from TGI specs. For more details, check out https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts.
class huggingface_hub.ChatCompletionOutputComplete
< source >( finish_reason: str index: int message: ChatCompletionOutputMessage logprobs: Optional = None )
class huggingface_hub.ChatCompletionOutputFunctionDefinition
< source >( arguments: Any name: str description: Optional = None )
class huggingface_hub.ChatCompletionOutputLogprob
< source >( logprob: float token: str top_logprobs: List )
class huggingface_hub.ChatCompletionOutputMessage
< source >( role: str content: Optional = None tool_calls: Optional = None )
class huggingface_hub.ChatCompletionOutputToolCall
< source >( function: ChatCompletionOutputFunctionDefinition id: str type: str )
class huggingface_hub.ChatCompletionOutputUsage
< source >( completion_tokens: int prompt_tokens: int total_tokens: int )
class huggingface_hub.ChatCompletionStreamOutput
< source >( choices: List created: int id: str model: str system_fingerprint: str )
Chat Completion Stream Output. Auto-generated from TGI specs. For more details, check out https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts.
class huggingface_hub.ChatCompletionStreamOutputChoice
< source >( delta: ChatCompletionStreamOutputDelta index: int finish_reason: Optional = None logprobs: Optional = None )
class huggingface_hub.ChatCompletionStreamOutputDelta
< source >( role: str content: Optional = None tool_calls: Optional = None )
class huggingface_hub.ChatCompletionStreamOutputDeltaToolCall
< source >( function: ChatCompletionStreamOutputFunction id: str index: int type: str )
class huggingface_hub.ChatCompletionStreamOutputFunction
< source >( arguments: str name: Optional = None )
class huggingface_hub.ChatCompletionStreamOutputLogprob
< source >( logprob: float token: str top_logprobs: List )
depth_estimation
Inputs for Depth Estimation inference
Outputs of inference for the Depth Estimation task
document_question_answering
class huggingface_hub.DocumentQuestionAnsweringInput
< source >( inputs: DocumentQuestionAnsweringInputData parameters: Optional = None )
Inputs for Document Question Answering inference
One (document, question) pair to answer
class huggingface_hub.DocumentQuestionAnsweringOutputElement
< source >( answer: str end: int score: float start: int words: List )
Outputs of inference for the Document Question Answering task
class huggingface_hub.DocumentQuestionAnsweringParameters
< source >( doc_stride: Optional = None handle_impossible_answer: Optional = None lang: Optional = None max_answer_len: Optional = None max_question_len: Optional = None max_seq_len: Optional = None top_k: Optional = None word_boxes: Optional = None )
Additional inference parameters Additional inference parameters for Document Question Answering
feature_extraction
class huggingface_hub.FeatureExtractionInput
< source >( inputs: str normalize: Optional = None prompt_name: Optional = None truncate: Optional = None truncation_direction: Optional = None )
Feature Extraction Input. Auto-generated from TEI specs. For more details, check out https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tei-import.ts.
fill_mask
Inputs for Fill Mask inference
class huggingface_hub.FillMaskOutputElement
< source >( score: float sequence: str token: int token_str: Any fill_mask_output_token_str: Optional = None )
Outputs of inference for the Fill Mask task
class huggingface_hub.FillMaskParameters
< source >( targets: Optional = None top_k: Optional = None )
Additional inference parameters Additional inference parameters for Fill Mask
image_classification
class huggingface_hub.ImageClassificationInput
< source >( inputs: Any parameters: Optional = None )
Inputs for Image Classification inference
Outputs of inference for the Image Classification task
class huggingface_hub.ImageClassificationParameters
< source >( function_to_apply: Optional = None top_k: Optional = None )
Additional inference parameters Additional inference parameters for Image Classification
image_segmentation
Inputs for Image Segmentation inference
class huggingface_hub.ImageSegmentationOutputElement
< source >( label: str mask: Any score: Optional = None )
Outputs of inference for the Image Segmentation task A predicted mask / segment
class huggingface_hub.ImageSegmentationParameters
< source >( mask_threshold: Optional = None overlap_mask_area_threshold: Optional = None subtask: Optional = None threshold: Optional = None )
Additional inference parameters Additional inference parameters for Image Segmentation
image_to_image
Inputs for Image To Image inference
Outputs of inference for the Image To Image task
class huggingface_hub.ImageToImageParameters
< source >( guidance_scale: Optional = None negative_prompt: Optional = None num_inference_steps: Optional = None target_size: Optional = None )
Additional inference parameters Additional inference parameters for Image To Image
The size in pixel of the output image
image_to_text
class huggingface_hub.ImageToTextGenerationParameters
< source >( do_sample: Optional = None early_stopping: Union = None epsilon_cutoff: Optional = None eta_cutoff: Optional = None max_length: Optional = None max_new_tokens: Optional = None min_length: Optional = None min_new_tokens: Optional = None num_beam_groups: Optional = None num_beams: Optional = None penalty_alpha: Optional = None temperature: Optional = None top_k: Optional = None top_p: Optional = None typical_p: Optional = None use_cache: Optional = None )
Parametrization of the text generation process Ad-hoc parametrization of the text generation process
Inputs for Image To Text inference
class huggingface_hub.ImageToTextOutput
< source >( generated_text: Any image_to_text_output_generated_text: Optional = None )
Outputs of inference for the Image To Text task
class huggingface_hub.ImageToTextParameters
< source >( generate: Optional = None max_new_tokens: Optional = None )
Additional inference parameters Additional inference parameters for Image To Text
object_detection
class huggingface_hub.ObjectDetectionBoundingBox
< source >( xmax: int xmin: int ymax: int ymin: int )
The predicted bounding box. Coordinates are relative to the top left corner of the input image.
Inputs for Object Detection inference
class huggingface_hub.ObjectDetectionOutputElement
< source >( box: ObjectDetectionBoundingBox label: str score: float )
Outputs of inference for the Object Detection task
Additional inference parameters Additional inference parameters for Object Detection
question_answering
class huggingface_hub.QuestionAnsweringInput
< source >( inputs: QuestionAnsweringInputData parameters: Optional = None )
Inputs for Question Answering inference
One (context, question) pair to answer
class huggingface_hub.QuestionAnsweringOutputElement
< source >( answer: str end: int score: float start: int )
Outputs of inference for the Question Answering task
class huggingface_hub.QuestionAnsweringParameters
< source >( align_to_words: Optional = None doc_stride: Optional = None handle_impossible_answer: Optional = None max_answer_len: Optional = None max_question_len: Optional = None max_seq_len: Optional = None top_k: Optional = None )
Additional inference parameters Additional inference parameters for Question Answering
sentence_similarity
class huggingface_hub.SentenceSimilarityInput
< source >( inputs: SentenceSimilarityInputData parameters: Optional = None )
Inputs for Sentence similarity inference
class huggingface_hub.SentenceSimilarityInputData
< source >( sentences: List source_sentence: str )
summarization
class huggingface_hub.SummarizationGenerationParameters
< source >( clean_up_tokenization_spaces: Optional = None generate_parameters: Optional = None truncation: Optional = None )
Additional inference parameters Additional inference parameters for Text2text Generation
Inputs for Summarization inference Inputs for Text2text Generation inference
Outputs of inference for the Summarization task
table_question_answering
class huggingface_hub.TableQuestionAnsweringInput
< source >( inputs: TableQuestionAnsweringInputData parameters: Optional = None )
Inputs for Table Question Answering inference
One (table, question) pair to answer
class huggingface_hub.TableQuestionAnsweringOutputElement
< source >( answer: str cells: List coordinates: List aggregator: Optional = None )
Outputs of inference for the Table Question Answering task
text2text_generation
class huggingface_hub.Text2TextGenerationInput
< source >( inputs: str parameters: Optional = None )
Inputs for Text2text Generation inference
class huggingface_hub.Text2TextGenerationOutput
< source >( generated_text: Any text2_text_generation_output_generated_text: Optional = None )
Outputs of inference for the Text2text Generation task
class huggingface_hub.Text2TextGenerationParameters
< source >( clean_up_tokenization_spaces: Optional = None generate_parameters: Optional = None truncation: Optional = None )
Additional inference parameters Additional inference parameters for Text2text Generation
text_classification
class huggingface_hub.TextClassificationInput
< source >( inputs: str parameters: Optional = None )
Inputs for Text Classification inference
Outputs of inference for the Text Classification task
class huggingface_hub.TextClassificationParameters
< source >( function_to_apply: Optional = None top_k: Optional = None )
Additional inference parameters Additional inference parameters for Text Classification
text_generation
class huggingface_hub.TextGenerationInput
< source >( inputs: str parameters: Optional = None stream: Optional = None )
Text Generation Input. Auto-generated from TGI specs. For more details, check out https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts.
class huggingface_hub.TextGenerationInputGenerateParameters
< source >( adapter_id: Optional = None best_of: Optional = None decoder_input_details: Optional = None details: Optional = None do_sample: Optional = None frequency_penalty: Optional = None grammar: Optional = None max_new_tokens: Optional = None repetition_penalty: Optional = None return_full_text: Optional = None seed: Optional = None stop: Optional = None temperature: Optional = None top_k: Optional = None top_n_tokens: Optional = None top_p: Optional = None truncate: Optional = None typical_p: Optional = None watermark: Optional = None )
class huggingface_hub.TextGenerationOutput
< source >( generated_text: str details: Optional = None )
Text Generation Output. Auto-generated from TGI specs. For more details, check out https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts.
class huggingface_hub.TextGenerationOutputBestOfSequence
< source >( finish_reason: TextGenerationOutputFinishReason generated_text: str generated_tokens: int prefill: List tokens: List seed: Optional = None top_tokens: Optional = None )
class huggingface_hub.TextGenerationOutputDetails
< source >( finish_reason: TextGenerationOutputFinishReason generated_tokens: int prefill: List tokens: List best_of_sequences: Optional = None seed: Optional = None top_tokens: Optional = None )
class huggingface_hub.TextGenerationOutputPrefillToken
< source >( id: int logprob: float text: str )
class huggingface_hub.TextGenerationOutputToken
< source >( id: int logprob: float special: bool text: str )
class huggingface_hub.TextGenerationStreamOutput
< source >( index: int token: TextGenerationStreamOutputToken details: Optional = None generated_text: Optional = None top_tokens: Optional = None )
Text Generation Stream Output. Auto-generated from TGI specs. For more details, check out https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-tgi-import.ts.
class huggingface_hub.TextGenerationStreamOutputStreamDetails
< source >( finish_reason: TextGenerationOutputFinishReason generated_tokens: int seed: Optional = None )
class huggingface_hub.TextGenerationStreamOutputToken
< source >( id: int logprob: float special: bool text: str )
text_to_audio
class huggingface_hub.TextToAudioGenerationParameters
< source >( do_sample: Optional = None early_stopping: Union = None epsilon_cutoff: Optional = None eta_cutoff: Optional = None max_length: Optional = None max_new_tokens: Optional = None min_length: Optional = None min_new_tokens: Optional = None num_beam_groups: Optional = None num_beams: Optional = None penalty_alpha: Optional = None temperature: Optional = None top_k: Optional = None top_p: Optional = None typical_p: Optional = None use_cache: Optional = None )
Parametrization of the text generation process Ad-hoc parametrization of the text generation process
Inputs for Text To Audio inference
class huggingface_hub.TextToAudioOutput
< source >( audio: Any sampling_rate: Any text_to_audio_output_sampling_rate: Optional = None )
Outputs of inference for the Text To Audio task
Additional inference parameters Additional inference parameters for Text To Audio
text_to_image
Inputs for Text To Image inference
Outputs of inference for the Text To Image task
class huggingface_hub.TextToImageParameters
< source >( guidance_scale: Optional = None negative_prompt: Optional = None num_inference_steps: Optional = None scheduler: Optional = None target_size: Optional = None )
Additional inference parameters Additional inference parameters for Text To Image
The size in pixel of the output image
token_classification
class huggingface_hub.TokenClassificationInput
< source >( inputs: str parameters: Optional = None )
Inputs for Token Classification inference
class huggingface_hub.TokenClassificationOutputElement
< source >( label: Any score: float end: Optional = None entity_group: Optional = None start: Optional = None word: Optional = None )
Outputs of inference for the Token Classification task
class huggingface_hub.TokenClassificationParameters
< source >( aggregation_strategy: Optional = None ignore_labels: Optional = None stride: Optional = None )
Additional inference parameters Additional inference parameters for Token Classification
translation
class huggingface_hub.TranslationGenerationParameters
< source >( clean_up_tokenization_spaces: Optional = None generate_parameters: Optional = None truncation: Optional = None )
Additional inference parameters Additional inference parameters for Text2text Generation
Inputs for Translation inference Inputs for Text2text Generation inference
Outputs of inference for the Translation task
video_classification
class huggingface_hub.VideoClassificationInput
< source >( inputs: Any parameters: Optional = None )
Inputs for Video Classification inference
Outputs of inference for the Video Classification task
class huggingface_hub.VideoClassificationParameters
< source >( frame_sampling_rate: Optional = None function_to_apply: Optional = None num_frames: Optional = None top_k: Optional = None )
Additional inference parameters Additional inference parameters for Video Classification
visual_question_answering
class huggingface_hub.VisualQuestionAnsweringInput
< source >( inputs: VisualQuestionAnsweringInputData parameters: Optional = None )
Inputs for Visual Question Answering inference
One (image, question) pair to answer
class huggingface_hub.VisualQuestionAnsweringOutputElement
< source >( label: Any score: float answer: Optional = None )
Outputs of inference for the Visual Question Answering task
Additional inference parameters Additional inference parameters for Visual Question Answering
zero_shot_classification
class huggingface_hub.ZeroShotClassificationInput
< source >( inputs: ZeroShotClassificationInputData parameters: Optional = None )
Inputs for Zero Shot Classification inference
class huggingface_hub.ZeroShotClassificationInputData
< source >( candidate_labels: List text: str )
The input text data, with candidate labels
Outputs of inference for the Zero Shot Classification task
class huggingface_hub.ZeroShotClassificationParameters
< source >( hypothesis_template: Optional = None multi_label: Optional = None )
Additional inference parameters Additional inference parameters for Zero Shot Classification
zero_shot_image_classification
class huggingface_hub.ZeroShotImageClassificationInput
< source >( inputs: ZeroShotImageClassificationInputData parameters: Optional = None )
Inputs for Zero Shot Image Classification inference
class huggingface_hub.ZeroShotImageClassificationInputData
< source >( candidate_labels: List image: Any )
The input image data, with candidate labels
class huggingface_hub.ZeroShotImageClassificationOutputElement
< source >( label: str score: float )
Outputs of inference for the Zero Shot Image Classification task
class huggingface_hub.ZeroShotImageClassificationParameters
< source >( hypothesis_template: Optional = None )
Additional inference parameters Additional inference parameters for Zero Shot Image Classification
zero_shot_object_detection
class huggingface_hub.ZeroShotObjectDetectionBoundingBox
< source >( xmax: int xmin: int ymax: int ymin: int )
The predicted bounding box. Coordinates are relative to the top left corner of the input image.
class huggingface_hub.ZeroShotObjectDetectionInput
< source >( inputs: ZeroShotObjectDetectionInputData parameters: Optional = None )
Inputs for Zero Shot Object Detection inference
class huggingface_hub.ZeroShotObjectDetectionInputData
< source >( candidate_labels: List image: Any )
The input image data, with candidate labels
class huggingface_hub.ZeroShotObjectDetectionOutputElement
< source >( box: ZeroShotObjectDetectionBoundingBox label: str score: float )
Outputs of inference for the Zero Shot Object Detection task