Update inference types (automated commit)

Author: Wauplin

docs/source/en/package_reference/inference_types.md

@@ -57,18 +57,12 @@ This part of the lib is still under development and will be improved in future r
 
 [[autodoc]] huggingface_hub.ChatCompletionInputFunctionName
 
-[[autodoc]] huggingface_hub.ChatCompletionInputJSONSchema
+[[autodoc]] huggingface_hub.ChatCompletionInputGrammarType
 
 [[autodoc]] huggingface_hub.ChatCompletionInputMessage
 
 [[autodoc]] huggingface_hub.ChatCompletionInputMessageChunk
 
-[[autodoc]] huggingface_hub.ChatCompletionInputResponseFormatJSONObject
-
-[[autodoc]] huggingface_hub.ChatCompletionInputResponseFormatJSONSchema
-
-[[autodoc]] huggingface_hub.ChatCompletionInputResponseFormatText
-
 [[autodoc]] huggingface_hub.ChatCompletionInputStreamOptions
 
 [[autodoc]] huggingface_hub.ChatCompletionInputTool
@@ -197,6 +191,18 @@ This part of the lib is still under development and will be improved in future r
 
 
 
+## image_to_video
+
+[[autodoc]] huggingface_hub.ImageToVideoInput
+
+[[autodoc]] huggingface_hub.ImageToVideoOutput
+
+[[autodoc]] huggingface_hub.ImageToVideoParameters
+
+[[autodoc]] huggingface_hub.ImageToVideoTargetSize
+
+
+
 ## object_detection
 
 [[autodoc]] huggingface_hub.ObjectDetectionBoundingBox

docs/source/ko/package_reference/inference_types.md

@@ -56,18 +56,12 @@ rendered properly in your Markdown viewer.
 
 [[autodoc]] huggingface_hub.ChatCompletionInputFunctionName
 
-[[autodoc]] huggingface_hub.ChatCompletionInputJSONSchema
+[[autodoc]] huggingface_hub.ChatCompletionInputGrammarType
 
 [[autodoc]] huggingface_hub.ChatCompletionInputMessage
 
 [[autodoc]] huggingface_hub.ChatCompletionInputMessageChunk
 
-[[autodoc]] huggingface_hub.ChatCompletionInputResponseFormatJSONObject
-
-[[autodoc]] huggingface_hub.ChatCompletionInputResponseFormatJSONSchema
-
-[[autodoc]] huggingface_hub.ChatCompletionInputResponseFormatText
-
 [[autodoc]] huggingface_hub.ChatCompletionInputStreamOptions
 
 [[autodoc]] huggingface_hub.ChatCompletionInputTool
@@ -196,6 +190,18 @@ rendered properly in your Markdown viewer.
 
 
 
+## image_to_video[[huggingface_hub.ImageToVideoInput]]
+
+[[autodoc]] huggingface_hub.ImageToVideoInput
+
+[[autodoc]] huggingface_hub.ImageToVideoOutput
+
+[[autodoc]] huggingface_hub.ImageToVideoParameters
+
+[[autodoc]] huggingface_hub.ImageToVideoTargetSize
+
+
+
 ## object_detection[[huggingface_hub.ObjectDetectionBoundingBox]]
 
 [[autodoc]] huggingface_hub.ObjectDetectionBoundingBox

src/huggingface_hub/inference/_generated/types/__init__.py

@@ -24,13 +24,10 @@
     ChatCompletionInputFunctionDefinition,
     ChatCompletionInputFunctionName,
     ChatCompletionInputGrammarType,
-    ChatCompletionInputJSONSchema,
+    ChatCompletionInputGrammarTypeType,
     ChatCompletionInputMessage,
     ChatCompletionInputMessageChunk,
     ChatCompletionInputMessageChunkType,
-    ChatCompletionInputResponseFormatJSONObject,
-    ChatCompletionInputResponseFormatJSONSchema,
-    ChatCompletionInputResponseFormatText,
     ChatCompletionInputStreamOptions,
     ChatCompletionInputTool,
     ChatCompletionInputToolCall,
@@ -85,6 +82,7 @@
     ImageToTextOutput,
     ImageToTextParameters,
 )
+from .image_to_video import ImageToVideoInput, ImageToVideoOutput, ImageToVideoParameters, ImageToVideoTargetSize
 from .object_detection import (
     ObjectDetectionBoundingBox,
     ObjectDetectionInput,

src/huggingface_hub/inference/_generated/types/chat_completion.py

@@ -3,7 +3,7 @@
 # See:
 #   - script: https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-codegen.ts
 #   - specs:  https://github.com/huggingface/huggingface.js/tree/main/packages/tasks/src/tasks.
-from typing import Any, Dict, List, Literal, Optional, Union
+from typing import Any, List, Literal, Optional, Union
 
 from .base import BaseInferenceType, dataclass_with_extra
 
@@ -45,51 +45,17 @@ class ChatCompletionInputMessage(BaseInferenceType):
     tool_calls: Optional[List[ChatCompletionInputToolCall]] = None
 
 
-@dataclass_with_extra
-class ChatCompletionInputJSONSchema(BaseInferenceType):
-    name: str
-    """
-    The name of the response format.
-    """
-    description: Optional[str] = None
-    """
-    A description of what the response format is for, used by the model to determine
-    how to respond in the format.
-    """
-    schema: Optional[Dict[str, object]] = None
-    """
-    The schema for the response format, described as a JSON Schema object. Learn how
-    to build JSON schemas [here](https://json-schema.org/).
-    """
-    strict: Optional[bool] = None
-    """
-    Whether to enable strict schema adherence when generating the output. If set to
-    true, the model will always follow the exact schema defined in the `schema`
-    field.
-    """
-
-
-@dataclass_with_extra
-class ChatCompletionInputResponseFormatText(BaseInferenceType):
-    type: Literal["text"]
-
-
-@dataclass_with_extra
-class ChatCompletionInputResponseFormatJSONSchema(BaseInferenceType):
-    type: Literal["json_schema"]
-    json_schema: ChatCompletionInputJSONSchema
+ChatCompletionInputGrammarTypeType = Literal["json", "regex", "json_schema"]
 
 
 @dataclass_with_extra
-class ChatCompletionInputResponseFormatJSONObject(BaseInferenceType):
-    type: Literal["json_object"]
-
-
-ChatCompletionInputGrammarType = Union[
-    ChatCompletionInputResponseFormatText,
-    ChatCompletionInputResponseFormatJSONSchema,
-    ChatCompletionInputResponseFormatJSONObject,
-]
+class ChatCompletionInputGrammarType(BaseInferenceType):
+    type: "ChatCompletionInputGrammarTypeType"
+    value: Any
+    """A string that represents a [JSON Schema](https://json-schema.org/).
+    JSON Schema is a declarative language that allows to annotate JSON documents
+    with types and descriptions.
+    """
 
 
 @dataclass_with_extra

src/huggingface_hub/inference/_generated/types/image_to_video.py

@@ -0,0 +1,60 @@
+# Inference code generated from the JSON schema spec in @huggingface/tasks.
+#
+# See:
+#   - script: https://github.com/huggingface/huggingface.js/blob/main/packages/tasks/scripts/inference-codegen.ts
+#   - specs:  https://github.com/huggingface/huggingface.js/tree/main/packages/tasks/src/tasks.
+from typing import Any, Optional
+
+from .base import BaseInferenceType, dataclass_with_extra
+
+
+@dataclass_with_extra
+class ImageToVideoTargetSize(BaseInferenceType):
+    """The size in pixel of the output video frames."""
+
+    height: int
+    width: int
+
+
+@dataclass_with_extra
+class ImageToVideoParameters(BaseInferenceType):
+    """Additional inference parameters for Image To Video"""
+
+    guidance_scale: Optional[float] = None
+    """For diffusion models. A higher guidance scale value encourages the model to generate
+    videos closely linked to the text prompt at the expense of lower image quality.
+    """
+    negative_prompt: Optional[str] = None
+    """One prompt to guide what NOT to include in video generation."""
+    num_frames: Optional[float] = None
+    """The num_frames parameter determines how many video frames are generated."""
+    num_inference_steps: Optional[int] = None
+    """The number of denoising steps. More denoising steps usually lead to a higher quality
+    video at the expense of slower inference.
+    """
+    prompt: Optional[str] = None
+    """The text prompt to guide the video generation."""
+    seed: Optional[int] = None
+    """Seed for the random number generator."""
+    target_size: Optional[ImageToVideoTargetSize] = None
+    """The size in pixel of the output video frames."""
+
+
+@dataclass_with_extra
+class ImageToVideoInput(BaseInferenceType):
+    """Inputs for Image To Video inference"""
+
+    inputs: str
+    """The input image data as a base64-encoded string. If no `parameters` are provided, you can
+    also provide the image data as a raw bytes payload.
+    """
+    parameters: Optional[ImageToVideoParameters] = None
+    """Additional inference parameters for Image To Video"""
+
+
+@dataclass_with_extra
+class ImageToVideoOutput(BaseInferenceType):
+    """Outputs of inference for the Image To Video task"""
+
+    video: Any
+    """The generated video returned as raw bytes in the payload."""