From 1d36e199f65ab0f3089964c5e884eb9bf12f18b0 Mon Sep 17 00:00:00 2001 From: Christian Byrne Date: Fri, 16 May 2025 14:51:40 -0700 Subject: [PATCH] Add OpenAI chat node (#188) --- comfy_api_nodes/apinode_utils.py | 17 + comfy_api_nodes/apis/__init__.py | 4263 +++++++++++++++--------------- comfy_api_nodes/nodes_openai.py | 524 +++- 3 files changed, 2659 insertions(+), 2145 deletions(-) diff --git a/comfy_api_nodes/apinode_utils.py b/comfy_api_nodes/apinode_utils.py index 210ce2719..07babe163 100644 --- a/comfy_api_nodes/apinode_utils.py +++ b/comfy_api_nodes/apinode_utils.py @@ -1,6 +1,7 @@ from __future__ import annotations import io import logging +import mimetypes from typing import Optional, Union from comfy.utils import common_upscale from comfy_api.input_impl import VideoFromFile @@ -318,6 +319,22 @@ def tensor_to_data_uri( return f"data:{mime_type};base64,{base64_string}" +def text_filepath_to_base64_string(filepath: str) -> str: + """Converts a text file to a base64 string.""" + with open(filepath, "rb") as f: + file_content = f.read() + return base64.b64encode(file_content).decode("utf-8") + + +def text_filepath_to_data_uri(filepath: str) -> str: + """Converts a text file to a data URI.""" + base64_string = text_filepath_to_base64_string(filepath) + mime_type, _ = mimetypes.guess_type(filepath) + if mime_type is None: + mime_type = "application/octet-stream" + return f"data:{mime_type};base64,{base64_string}" + + def upload_file_to_comfyapi( file_bytes_io: BytesIO, filename: str, diff --git a/comfy_api_nodes/apis/__init__.py b/comfy_api_nodes/apis/__init__.py index 18e6d2487..3f07baade 100644 --- a/comfy_api_nodes/apis/__init__.py +++ b/comfy_api_nodes/apis/__init__.py @@ -1,67 +1,197 @@ # generated by datamodel-codegen: # filename: filtered-openapi.yaml -# timestamp: 2025-05-04T04:12:39+00:00 +# timestamp: 2025-05-15T19:13:25+00:00 from __future__ import annotations -from datetime import datetime +from datetime import date, datetime from enum import Enum from typing import Any, Dict, List, Literal, Optional, Union from uuid import UUID -from pydantic import AnyUrl, BaseModel, Field, RootModel, StrictBytes +from pydantic import AnyUrl, BaseModel, ConfigDict, Field, RootModel, StrictBytes -class PersonalAccessToken(BaseModel): - id: Optional[UUID] = Field(None, description='Unique identifier for the GitCommit') - name: Optional[str] = Field( - None, - description='Required. The name of the token. Can be a simple description.', - ) - description: Optional[str] = Field( - None, - description="Optional. A more detailed description of the token's intended use.", +class APIKey(BaseModel): + created_at: Optional[datetime] = None + description: Optional[str] = None + id: Optional[str] = None + key_prefix: Optional[str] = None + name: Optional[str] = None + + +class APIKeyWithPlaintext(APIKey): + plaintext_key: Optional[str] = Field( + None, description='The full API key (only returned at creation)' ) + + +class AuditLog(BaseModel): createdAt: Optional[datetime] = Field( - None, description='[Output Only]The date and time the token was created.' + None, description='The date and time the event was created' ) - token: Optional[str] = Field( + event_id: Optional[str] = Field(None, description='the id of the event') + event_type: Optional[str] = Field(None, description='the type of the event') + params: Optional[Dict[str, Any]] = Field( + None, description='data related to the event' + ) + + +class OutputFormat(str, Enum): + jpeg = 'jpeg' + png = 'png' + + +class BFLFluxPro11GenerateRequest(BaseModel): + height: int = Field(..., description='Height of the generated image') + image_prompt: Optional[str] = Field(None, description='Optional image prompt') + output_format: Optional[OutputFormat] = Field( + None, description='Output image format' + ) + prompt: str = Field(..., description='The main text prompt for image generation') + prompt_upsampling: Optional[bool] = Field( + None, description='Whether to use prompt upsampling' + ) + safety_tolerance: Optional[int] = Field(None, description='Safety tolerance level') + seed: Optional[int] = Field(None, description='Random seed for reproducibility') + webhook_secret: Optional[str] = Field( + None, description='Optional webhook secret for async processing' + ) + webhook_url: Optional[str] = Field( + None, description='Optional webhook URL for async processing' + ) + width: int = Field(..., description='Width of the generated image') + + +class BFLFluxPro11GenerateResponse(BaseModel): + id: str = Field(..., description='Job ID for tracking') + polling_url: str = Field(..., description='URL to poll for results') + + +class BFLFluxProGenerateRequest(BaseModel): + guidance_scale: Optional[float] = Field( + None, description='The guidance scale for generation.', ge=1.0, le=20.0 + ) + height: int = Field( + ..., description='The height of the image to generate.', ge=64, le=2048 + ) + negative_prompt: Optional[str] = Field( + None, description='The negative prompt for image generation.' + ) + num_images: Optional[int] = Field( + None, description='The number of images to generate.', ge=1, le=4 + ) + num_inference_steps: Optional[int] = Field( + None, description='The number of inference steps.', ge=1, le=100 + ) + prompt: str = Field(..., description='The text prompt for image generation.') + seed: Optional[int] = Field(None, description='The seed value for reproducibility.') + width: int = Field( + ..., description='The width of the image to generate.', ge=64, le=2048 + ) + + +class BFLFluxProGenerateResponse(BaseModel): + id: str = Field(..., description='The unique identifier for the generation task.') + polling_url: str = Field(..., description='URL to poll for the generation result.') + + +class Status(str, Enum): + in_progress = 'in_progress' + completed = 'completed' + incomplete = 'incomplete' + + +class Type(str, Enum): + computer_call = 'computer_call' + + +class ComputerToolCall(BaseModel): + action: Dict[str, Any] + call_id: str = Field( + ..., + description='An identifier used when responding to the tool call with output.\n', + ) + id: str = Field(..., description='The unique ID of the computer call.') + status: Status = Field( + ..., + description='The status of the item. One of `in_progress`, `completed`, or\n`incomplete`. Populated when items are returned via API.\n', + ) + type: Type = Field( + ..., description='The type of the computer call. Always `computer_call`.' + ) + + +class Environment(str, Enum): + windows = 'windows' + mac = 'mac' + linux = 'linux' + ubuntu = 'ubuntu' + browser = 'browser' + + +class Type1(str, Enum): + computer_use_preview = 'computer_use_preview' + + +class ComputerUsePreviewTool(BaseModel): + display_height: int = Field(..., description='The height of the computer display.') + display_width: int = Field(..., description='The width of the computer display.') + environment: Environment = Field( + ..., description='The type of computer environment to control.' + ) + type: Literal['ComputerUsePreviewTool'] = Field( + ..., + description='The type of the computer use tool. Always `computer_use_preview`.', + ) + + +class CreateAPIKeyRequest(BaseModel): + description: Optional[str] = None + name: str + + +class Customer(BaseModel): + createdAt: Optional[datetime] = Field( + None, description='The date and time the user was created' + ) + email: Optional[str] = Field(None, description='The email address for this user') + id: str = Field(..., description='The firebase UID of the user') + is_admin: Optional[bool] = Field(None, description='Whether the user is an admin') + metronome_id: Optional[str] = Field(None, description='The Metronome customer ID') + name: Optional[str] = Field(None, description='The name for this user') + stripe_id: Optional[str] = Field(None, description='The Stripe customer ID') + updatedAt: Optional[datetime] = Field( + None, description='The date and time the user was last updated' + ) + + +class CustomerStorageResourceResponse(BaseModel): + download_url: Optional[str] = Field( None, - description='[Output Only]. The personal access token. Only returned during creation.', + description='The signed URL to use for downloading the file from the specified path', + ) + existing_file: Optional[bool] = Field( + None, description='Whether an existing file with the same hash was found' + ) + expires_at: Optional[datetime] = Field( + None, description='When the signed URL will expire' + ) + upload_url: Optional[str] = Field( + None, + description='The signed URL to use for uploading the file to the specified path', ) -class GitCommitSummary(BaseModel): - commit_hash: Optional[str] = Field(None, description='The hash of the commit') - commit_name: Optional[str] = Field(None, description='The name of the commit') - branch_name: Optional[str] = Field( - None, description='The branch where the commit was made' - ) - author: Optional[str] = Field(None, description='The author of the commit') - timestamp: Optional[datetime] = Field( - None, description='The timestamp when the commit was made' - ) - status_summary: Optional[Dict[str, str]] = Field( - None, description='A map of operating system to status pairs' - ) +class Role(str, Enum): + user = 'user' + assistant = 'assistant' + system = 'system' + developer = 'developer' -class User(BaseModel): - id: Optional[str] = Field(None, description='The unique id for this user.') - email: Optional[str] = Field(None, description='The email address for this user.') - name: Optional[str] = Field(None, description='The name for this user.') - isApproved: Optional[bool] = Field( - None, description='Indicates if the user is approved.' - ) - isAdmin: Optional[bool] = Field( - None, description='Indicates if the user has admin privileges.' - ) - - -class PublisherUser(BaseModel): - id: Optional[str] = Field(None, description='The unique id for this user.') - email: Optional[str] = Field(None, description='The email address for this user.') - name: Optional[str] = Field(None, description='The name for this user.') +class Type2(str, Enum): + message = 'message' class ComfyNode(BaseModel): @@ -204,168 +334,156 @@ class ErrorResponse(BaseModel): message: str -class StorageFile(BaseModel): - id: Optional[UUID] = Field( - None, description='Unique identifier for the storage file' - ) - file_path: Optional[str] = Field(None, description='Path to the file in storage') - public_url: Optional[str] = Field(None, description='Public URL') +class Type3(str, Enum): + file_search = 'file_search' -class PublisherMember(BaseModel): - id: Optional[str] = Field( - None, description='The unique identifier for the publisher member.' - ) - user: Optional[PublisherUser] = Field( - None, description='The user associated with this publisher member.' - ) - role: Optional[str] = Field( - None, description='The role of the user in the publisher.' +class FileSearchTool(BaseModel): + type: Literal['FileSearchTool'] = Field(..., description='The type of tool') + vector_store_ids: List[str] = Field( + ..., description='IDs of vector stores to search in' ) -class ComfyNode(BaseModel): - comfy_node_name: Optional[str] = Field( - None, description='Unique identifier for the node' +class Result(BaseModel): + file_id: Optional[str] = Field(None, description='The unique ID of the file.\n') + filename: Optional[str] = Field(None, description='The name of the file.\n') + score: Optional[float] = Field( + None, description='The relevance score of the file - a value between 0 and 1.\n' ) - category: Optional[str] = Field( - None, - description='UI category where the node is listed, used for grouping nodes.', + text: Optional[str] = Field( + None, description='The text that was retrieved from the file.\n' ) + + +class Status1(str, Enum): + in_progress = 'in_progress' + searching = 'searching' + completed = 'completed' + incomplete = 'incomplete' + failed = 'failed' + + +class Type4(str, Enum): + file_search_call = 'file_search_call' + + +class FileSearchToolCall(BaseModel): + id: str = Field(..., description='The unique ID of the file search tool call.\n') + queries: List[str] = Field( + ..., description='The queries used to search for files.\n' + ) + results: Optional[List[Result]] = Field( + None, description='The results of the file search tool call.\n' + ) + status: Status1 = Field( + ..., + description='The status of the file search tool call. One of `in_progress`, \n`searching`, `incomplete` or `failed`,\n', + ) + type: Type4 = Field( + ..., + description='The type of the file search tool call. Always `file_search_call`.\n', + ) + + +class Type5(str, Enum): + function = 'function' + + +class FunctionTool(BaseModel): description: Optional[str] = Field( - None, description="Brief description of the node's functionality or purpose." + None, description='Description of what the function does' ) - input_types: Optional[str] = Field(None, description='Defines input parameters') - deprecated: Optional[bool] = Field( + name: str = Field(..., description='Name of the function') + parameters: Dict[str, Any] = Field( + ..., description='JSON Schema object describing the function parameters' + ) + type: Literal['FunctionTool'] = Field(..., description='The type of tool') + + +class Status2(str, Enum): + in_progress = 'in_progress' + completed = 'completed' + incomplete = 'incomplete' + + +class Type6(str, Enum): + function_call = 'function_call' + + +class FunctionToolCall(BaseModel): + arguments: str = Field( + ..., description='A JSON string of the arguments to pass to the function.\n' + ) + call_id: str = Field( + ..., + description='The unique ID of the function tool call generated by the model.\n', + ) + id: Optional[str] = Field( + None, description='The unique ID of the function tool call.\n' + ) + name: str = Field(..., description='The name of the function to run.\n') + status: Optional[Status2] = Field( None, - description='Indicates if the node is deprecated. Deprecated nodes are hidden in the UI.', + description='The status of the item. One of `in_progress`, `completed`, or\n`incomplete`. Populated when items are returned via API.\n', ) - experimental: Optional[bool] = Field( - None, - description='Indicates if the node is experimental, subject to changes or removal.', - ) - output_is_list: Optional[List[bool]] = Field( - None, description='Boolean values indicating if each output is a list.' - ) - return_names: Optional[str] = Field( - None, description='Names of the outputs for clarity in workflows.' - ) - return_types: Optional[str] = Field( - None, description='Specifies the types of outputs produced by the node.' - ) - function: Optional[str] = Field( - None, description='Name of the entry-point function to execute the node.' + type: Type6 = Field( + ..., description='The type of the function tool call. Always `function_call`.\n' ) -class ComfyNodeCloudBuildInfo(BaseModel): - project_id: Optional[str] = None - project_number: Optional[str] = None - location: Optional[str] = None - build_id: Optional[str] = None +class GeminiCitation(BaseModel): + authors: Optional[List[str]] = None + endIndex: Optional[int] = None + license: Optional[str] = None + publicationDate: Optional[date] = None + startIndex: Optional[int] = None + title: Optional[str] = None + uri: Optional[str] = None -class Error(BaseModel): - message: Optional[str] = Field( - None, description='A clear and concise description of the error.' - ) - details: Optional[List[str]] = Field( - None, - description='Optional detailed information about the error or hints for resolving it.', +class GeminiCitationMetadata(BaseModel): + citations: Optional[List[GeminiCitation]] = None + + +class GeminiFunctionDeclaration(BaseModel): + description: Optional[str] = None + name: str + parameters: Dict[str, Any] = Field( + ..., description='JSON schema for the function parameters' ) -class NodeVersionUpdateRequest(BaseModel): - changelog: Optional[str] = Field( - None, description='The changelog describing the version changes.' - ) - deprecated: Optional[bool] = Field( - None, description='Whether the version is deprecated.' - ) +class GeminiGenerationConfig(BaseModel): + maxOutputTokens: Optional[int] = None + stopSequences: Optional[List[str]] = None + temperature: Optional[float] = None + topK: Optional[int] = None + topP: Optional[float] = None -class NodeStatus(str, Enum): - NodeStatusActive = 'NodeStatusActive' - NodeStatusDeleted = 'NodeStatusDeleted' - NodeStatusBanned = 'NodeStatusBanned' +class GeminiInlineData(BaseModel): + data: Optional[str] = None + mimeType: Optional[str] = None -class NodeVersionStatus(str, Enum): - NodeVersionStatusActive = 'NodeVersionStatusActive' - NodeVersionStatusDeleted = 'NodeVersionStatusDeleted' - NodeVersionStatusBanned = 'NodeVersionStatusBanned' - NodeVersionStatusPending = 'NodeVersionStatusPending' - NodeVersionStatusFlagged = 'NodeVersionStatusFlagged' +class GeminiPart(BaseModel): + inlineData: Optional[GeminiInlineData] = None + text: Optional[str] = None -class PublisherStatus(str, Enum): - PublisherStatusActive = 'PublisherStatusActive' - PublisherStatusBanned = 'PublisherStatusBanned' +class GeminiSafetyRating(BaseModel): + category: Optional[str] = None + probability: Optional[str] = None -class WorkflowRunStatus(str, Enum): - WorkflowRunStatusStarted = 'WorkflowRunStatusStarted' - WorkflowRunStatusFailed = 'WorkflowRunStatusFailed' - WorkflowRunStatusCompleted = 'WorkflowRunStatusCompleted' +class GeminiSafetySetting(BaseModel): + category: str + threshold: str -class MachineStats(BaseModel): - machine_name: Optional[str] = Field(None, description='Name of the machine.') - os_version: Optional[str] = Field( - None, description='The operating system version. eg. Ubuntu Linux 20.04' - ) - gpu_type: Optional[str] = Field( - None, description='The GPU type. eg. NVIDIA Tesla K80' - ) - cpu_capacity: Optional[str] = Field(None, description='Total CPU on the machine.') - initial_cpu: Optional[str] = Field( - None, description='Initial CPU available before the job starts.' - ) - memory_capacity: Optional[str] = Field( - None, description='Total memory on the machine.' - ) - initial_ram: Optional[str] = Field( - None, description='Initial RAM available before the job starts.' - ) - vram_time_series: Optional[Dict[str, Any]] = Field( - None, description='Time series of VRAM usage.' - ) - disk_capacity: Optional[str] = Field( - None, description='Total disk capacity on the machine.' - ) - initial_disk: Optional[str] = Field( - None, description='Initial disk available before the job starts.' - ) - pip_freeze: Optional[str] = Field(None, description='The pip freeze output') - - -class Customer(BaseModel): - id: str = Field(..., description='The firebase UID of the user') - email: Optional[str] = Field(None, description='The email address for this user') - name: Optional[str] = Field(None, description='The name for this user') - createdAt: Optional[datetime] = Field( - None, description='The date and time the user was created' - ) - updatedAt: Optional[datetime] = Field( - None, description='The date and time the user was last updated' - ) - - -class MagicPrompt(str, Enum): - ON = 'ON' - OFF = 'OFF' - - -class ColorPalette(BaseModel): - name: str = Field(..., description='Name of the color palette', examples=['PASTEL']) - - -class StyleCode(RootModel[str]): - root: str = Field(..., pattern='^[0-9A-Fa-f]{8}$') - - -class StyleType(str, Enum): - GENERAL = 'GENERAL' +class GeminiTool(BaseModel): + functionDeclarations: Optional[List[GeminiFunctionDeclaration]] = None class IdeogramColorPalette1(BaseModel): @@ -397,27 +515,17 @@ class IdeogramColorPalette( class ImageRequest(BaseModel): - prompt: str = Field( - ..., description='Required. The prompt to use to generate the image.' - ) aspect_ratio: Optional[str] = Field( None, description="Optional. The aspect ratio (e.g., 'ASPECT_16_9', 'ASPECT_1_1'). Cannot be used with resolution. Defaults to 'ASPECT_1_1' if unspecified.", ) - model: str = Field(..., description="The model used (e.g., 'V_2', 'V_2A_TURBO')") + color_palette: Optional[Dict[str, Any]] = Field( + None, description='Optional. Color palette object. Only for V_2, V_2_TURBO.' + ) magic_prompt_option: Optional[str] = Field( None, description="Optional. MagicPrompt usage ('AUTO', 'ON', 'OFF')." ) - seed: Optional[int] = Field( - None, - description='Optional. A number between 0 and 2147483647.', - ge=0, - le=2147483647, - ) - style_type: Optional[str] = Field( - None, - description="Optional. Style type ('AUTO', 'GENERAL', 'REALISTIC', 'DESIGN', 'RENDER_3D', 'ANIME'). Only for models V_2 and above.", - ) + model: str = Field(..., description="The model used (e.g., 'V_2', 'V_2A_TURBO')") negative_prompt: Optional[str] = Field( None, description='Optional. Description of what to exclude. Only for V_1, V_1_TURBO, V_2, V_2_TURBO.', @@ -428,12 +536,22 @@ class ImageRequest(BaseModel): ge=1, le=8, ) + prompt: str = Field( + ..., description='Required. The prompt to use to generate the image.' + ) resolution: Optional[str] = Field( None, description="Optional. Resolution (e.g., 'RESOLUTION_1024_1024'). Only for model V_2. Cannot be used with aspect_ratio.", ) - color_palette: Optional[Dict[str, Any]] = Field( - None, description='Optional. Color palette object. Only for V_2, V_2_TURBO.' + seed: Optional[int] = Field( + None, + description='Optional. A number between 0 and 2147483647.', + ge=0, + le=2147483647, + ) + style_type: Optional[str] = Field( + None, + description="Optional. Style type ('AUTO', 'GENERAL', 'REALISTIC', 'DESIGN', 'RENDER_3D', 'ANIME'). Only for models V_2 and above.", ) @@ -450,23 +568,23 @@ class IdeogramGenerateRequest(BaseModel): class Datum(BaseModel): + is_image_safe: Optional[bool] = Field( + None, description='Indicates whether the image is considered safe.' + ) prompt: Optional[str] = Field( None, description='The prompt used to generate this image.' ) resolution: Optional[str] = Field( None, description="The resolution of the generated image (e.g., '1024x1024')." ) - is_image_safe: Optional[bool] = Field( - None, description='Indicates whether the image is considered safe.' - ) seed: Optional[int] = Field( None, description='The seed value used for this generation.' ) - url: Optional[str] = Field(None, description='URL to the generated image.') style_type: Optional[str] = Field( None, description="The style type used for generation (e.g., 'REALISTIC', 'ANIME').", ) + url: Optional[str] = Field(None, description='URL to the generated image.') class IdeogramGenerateResponse(BaseModel): @@ -478,49 +596,17 @@ class IdeogramGenerateResponse(BaseModel): ) -class RenderingSpeed1(str, Enum): - TURBO = 'TURBO' - DEFAULT = 'DEFAULT' - QUALITY = 'QUALITY' - - -class MagicPrompt1(str, Enum): - AUTO = 'AUTO' - ON = 'ON' - OFF = 'OFF' - - -class StyleType1(str, Enum): - AUTO = 'AUTO' - GENERAL = 'GENERAL' - REALISTIC = 'REALISTIC' - DESIGN = 'DESIGN' - - -class IdeogramV3RemixRequest(BaseModel): - image: Optional[StrictBytes] = None - prompt: str - image_weight: Optional[int] = Field(50, ge=1, le=100) - seed: Optional[int] = Field(None, ge=0, le=2147483647) - resolution: Optional[str] = None - aspect_ratio: Optional[str] = None - rendering_speed: Optional[RenderingSpeed1] = None - magic_prompt: Optional[MagicPrompt1] = None - negative_prompt: Optional[str] = None - num_images: Optional[int] = Field(None, ge=1, le=8) - color_palette: Optional[Dict[str, Any]] = None - style_codes: Optional[List[str]] = None - style_type: Optional[StyleType1] = None - style_reference_images: Optional[List[StrictBytes]] = None +class StyleCode(RootModel[str]): + root: str = Field(..., pattern='^[0-9A-Fa-f]{8}$') class Datum1(BaseModel): + is_image_safe: Optional[bool] = None prompt: Optional[str] = None resolution: Optional[str] = None - is_image_safe: Optional[bool] = None seed: Optional[int] = None - url: Optional[str] = None style_type: Optional[str] = None + url: Optional[str] = None class IdeogramV3IdeogramResponse(BaseModel): @@ -528,74 +614,201 @@ class IdeogramV3IdeogramResponse(BaseModel): data: Optional[List[Datum1]] = None +class RenderingSpeed1(str, Enum): + TURBO = 'TURBO' + DEFAULT = 'DEFAULT' + QUALITY = 'QUALITY' + + class IdeogramV3ReframeRequest(BaseModel): - image: Optional[StrictBytes] = None - resolution: str - num_images: Optional[int] = Field(None, ge=1, le=8) - seed: Optional[int] = Field(None, ge=0, le=2147483647) - rendering_speed: Optional[RenderingSpeed1] = None color_palette: Optional[Dict[str, Any]] = None + image: Optional[StrictBytes] = None + num_images: Optional[int] = Field(None, ge=1, le=8) + rendering_speed: Optional[RenderingSpeed1] = None + resolution: str + seed: Optional[int] = Field(None, ge=0, le=2147483647) style_codes: Optional[List[str]] = None style_reference_images: Optional[List[StrictBytes]] = None +class MagicPrompt(str, Enum): + AUTO = 'AUTO' + ON = 'ON' + OFF = 'OFF' + + +class StyleType(str, Enum): + AUTO = 'AUTO' + GENERAL = 'GENERAL' + REALISTIC = 'REALISTIC' + DESIGN = 'DESIGN' + + +class IdeogramV3RemixRequest(BaseModel): + aspect_ratio: Optional[str] = None + color_palette: Optional[Dict[str, Any]] = None + image: Optional[StrictBytes] = None + image_weight: Optional[int] = Field(50, ge=1, le=100) + magic_prompt: Optional[MagicPrompt] = None + negative_prompt: Optional[str] = None + num_images: Optional[int] = Field(None, ge=1, le=8) + prompt: str + rendering_speed: Optional[RenderingSpeed1] = None + resolution: Optional[str] = None + seed: Optional[int] = Field(None, ge=0, le=2147483647) + style_codes: Optional[List[str]] = None + style_reference_images: Optional[List[StrictBytes]] = None + style_type: Optional[StyleType] = None + + class IdeogramV3ReplaceBackgroundRequest(BaseModel): - image: Optional[StrictBytes] = None - prompt: str - magic_prompt: Optional[MagicPrompt1] = None - num_images: Optional[int] = Field(None, ge=1, le=8) - seed: Optional[int] = Field(None, ge=0, le=2147483647) - rendering_speed: Optional[RenderingSpeed1] = None color_palette: Optional[Dict[str, Any]] = None + image: Optional[StrictBytes] = None + magic_prompt: Optional[MagicPrompt] = None + num_images: Optional[int] = Field(None, ge=1, le=8) + prompt: str + rendering_speed: Optional[RenderingSpeed1] = None + seed: Optional[int] = Field(None, ge=0, le=2147483647) style_codes: Optional[List[str]] = None style_reference_images: Optional[List[StrictBytes]] = None -class KlingTaskStatus(str, Enum): - submitted = 'submitted' - processing = 'processing' - succeed = 'succeed' - failed = 'failed' +class ColorPalette(BaseModel): + name: str = Field(..., description='Name of the color palette', examples=['PASTEL']) -class KlingVideoGenModelName(str, Enum): - kling_v1 = 'kling-v1' - kling_v1_5 = 'kling-v1-5' - kling_v1_6 = 'kling-v1-6' - kling_v2_master = 'kling-v2-master' +class MagicPrompt2(str, Enum): + ON = 'ON' + OFF = 'OFF' -class KlingVideoGenMode(str, Enum): - std = 'std' - pro = 'pro' +class StyleType1(str, Enum): + GENERAL = 'GENERAL' -class KlingVideoGenAspectRatio(str, Enum): - field_16_9 = '16:9' - field_9_16 = '9:16' +class ImagenImageGenerationInstance(BaseModel): + prompt: str = Field(..., description='Text prompt for image generation') + + +class AspectRatio(str, Enum): field_1_1 = '1:1' + field_9_16 = '9:16' + field_16_9 = '16:9' + field_3_4 = '3:4' + field_4_3 = '4:3' -class KlingVideoGenDuration(str, Enum): - field_5 = '5' - field_10 = '10' +class PersonGeneration(str, Enum): + dont_allow = 'dont_allow' + allow_adult = 'allow_adult' + allow_all = 'allow_all' -class KlingVideoGenCfgScale(RootModel[float]): - root: float = Field( - ..., - description="Flexibility in video generation. The higher the value, the lower the model's degree of flexibility, and the stronger the relevance to the user's prompt.", - ge=0.0, - le=1.0, +class SafetySetting(str, Enum): + block_most = 'block_most' + block_some = 'block_some' + block_few = 'block_few' + block_fewest = 'block_fewest' + + +class ImagenImagePrediction(BaseModel): + bytesBase64Encoded: Optional[str] = Field( + None, description='Base64-encoded image content' + ) + mimeType: Optional[str] = Field( + None, description='MIME type of the generated image' + ) + prompt: Optional[str] = Field( + None, description='Enhanced or rewritten prompt used to generate this image' ) -class KlingCameraControlType(str, Enum): - simple = 'simple' - down_back = 'down_back' - forward_up = 'forward_up' - right_turn_forward = 'right_turn_forward' - left_turn_forward = 'left_turn_forward' +class MimeType(str, Enum): + image_png = 'image/png' + image_jpeg = 'image/jpeg' + + +class ImagenOutputOptions(BaseModel): + compressionQuality: Optional[int] = Field(None, ge=0, le=100) + mimeType: Optional[MimeType] = None + + +class Includable(str, Enum): + file_search_call_results = 'file_search_call.results' + message_input_image_image_url = 'message.input_image.image_url' + computer_call_output_output_image_url = 'computer_call_output.output.image_url' + + +class Type7(str, Enum): + input_file = 'input_file' + + +class InputFileContent(BaseModel): + file_data: Optional[str] = Field( + None, description='The content of the file to be sent to the model.\n' + ) + file_id: Optional[str] = Field( + None, description='The ID of the file to be sent to the model.' + ) + filename: Optional[str] = Field( + None, description='The name of the file to be sent to the model.' + ) + type: Type7 = Field( + ..., description='The type of the input item. Always `input_file`.' + ) + + +class Detail(str, Enum): + low = 'low' + high = 'high' + auto = 'auto' + + +class Type8(str, Enum): + input_image = 'input_image' + + +class InputImageContent(BaseModel): + detail: Detail = Field( + ..., + description='The detail level of the image to be sent to the model. One of `high`, `low`, or `auto`. Defaults to `auto`.', + ) + file_id: Optional[str] = Field( + None, description='The ID of the file to be sent to the model.' + ) + image_url: Optional[str] = Field( + None, + description='The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL.', + ) + type: Type8 = Field( + ..., description='The type of the input item. Always `input_image`.' + ) + + +class Role1(str, Enum): + user = 'user' + system = 'system' + developer = 'developer' + + +class Type9(str, Enum): + message = 'message' + + +class Type10(str, Enum): + input_text = 'input_text' + + +class InputTextContent(BaseModel): + text: str = Field(..., description='The text input to the model.') + type: Type10 = Field( + ..., description='The type of the input item. Always `input_text`.' + ) + + +class KlingAudioUploadType(str, Enum): + file = 'file' + url = 'url' class KlingCameraConfig(BaseModel): @@ -605,27 +818,27 @@ class KlingCameraConfig(BaseModel): ge=-10.0, le=10.0, ) - vertical: Optional[float] = Field( - None, - description="Controls camera's movement along vertical axis (y-axis). Negative indicates downward, positive indicates upward.", - ge=-10.0, - le=10.0, - ) pan: Optional[float] = Field( None, description="Controls camera's rotation in vertical plane (x-axis). Negative indicates downward rotation, positive indicates upward rotation.", ge=-10.0, le=10.0, ) + roll: Optional[float] = Field( + None, + description="Controls camera's rolling amount (z-axis). Negative indicates counterclockwise, positive indicates clockwise.", + ge=-10.0, + le=10.0, + ) tilt: Optional[float] = Field( None, description="Controls camera's rotation in horizontal plane (y-axis). Negative indicates left rotation, positive indicates right rotation.", ge=-10.0, le=10.0, ) - roll: Optional[float] = Field( + vertical: Optional[float] = Field( None, - description="Controls camera's rolling amount (z-axis). Negative indicates counterclockwise, positive indicates clockwise.", + description="Controls camera's movement along vertical axis (y-axis). Negative indicates downward, positive indicates upward.", ge=-10.0, le=10.0, ) @@ -637,39 +850,12 @@ class KlingCameraConfig(BaseModel): ) -class KlingVideoResult(BaseModel): - id: Optional[str] = Field(None, description='Generated video ID') - url: Optional[AnyUrl] = Field(None, description='URL for generated video') - duration: Optional[str] = Field(None, description='Total video duration') - - -class KlingAudioUploadType(str, Enum): - file = 'file' - url = 'url' - - -class KlingLipSyncMode(str, Enum): - text2video = 'text2video' - audio2video = 'audio2video' - - -class KlingLipSyncVoiceLanguage(str, Enum): - zh = 'zh' - en = 'en' - - -class KlingDualCharacterEffectsScene(str, Enum): - hug = 'hug' - kiss = 'kiss' - heart_gesture = 'heart_gesture' - - -class KlingSingleImageEffectsScene(str, Enum): - bloombloom = 'bloombloom' - dizzydizzy = 'dizzydizzy' - fuzzyfuzzy = 'fuzzyfuzzy' - squish = 'squish' - expansion = 'expansion' +class KlingCameraControlType(str, Enum): + simple = 'simple' + down_back = 'down_back' + forward_up = 'forward_up' + right_turn_forward = 'right_turn_forward' + left_turn_forward = 'left_turn_forward' class KlingCharacterEffectModelName(str, Enum): @@ -678,18 +864,50 @@ class KlingCharacterEffectModelName(str, Enum): kling_v1_6 = 'kling-v1-6' -class KlingSingleImageEffectModelName(str, Enum): - kling_v1_6 = 'kling-v1-6' - - -class KlingSingleImageEffectDuration(str, Enum): - field_5 = '5' +class KlingDualCharacterEffectsScene(str, Enum): + hug = 'hug' + kiss = 'kiss' + heart_gesture = 'heart_gesture' class KlingDualCharacterImages(RootModel[List[str]]): root: List[str] = Field(..., max_length=2, min_length=2) +class KlingErrorResponse(BaseModel): + code: int = Field( + ..., + description='- 1000: Authentication failed\n- 1001: Authorization is empty\n- 1002: Authorization is invalid\n- 1003: Authorization is not yet valid\n- 1004: Authorization has expired\n- 1100: Account exception\n- 1101: Account in arrears (postpaid scenario)\n- 1102: Resource pack depleted or expired (prepaid scenario)\n- 1103: Unauthorized access to requested resource\n- 1200: Invalid request parameters\n- 1201: Invalid parameters\n- 1202: Invalid request method\n- 1203: Requested resource does not exist\n- 1300: Trigger platform strategy\n- 1301: Trigger content security policy\n- 1302: API request too frequent\n- 1303: Concurrency/QPS exceeds limit\n- 1304: Trigger IP whitelist policy\n- 5000: Internal server error\n- 5001: Service temporarily unavailable\n- 5002: Server internal timeout\n', + ) + message: str = Field(..., description='Human-readable error message') + request_id: str = Field( + ..., description='Request ID for tracking and troubleshooting' + ) + + +class Trajectory(BaseModel): + x: Optional[int] = Field( + None, + description='The horizontal coordinate of trajectory point. Based on bottom-left corner of image as origin (0,0).', + ) + y: Optional[int] = Field( + None, + description='The vertical coordinate of trajectory point. Based on bottom-left corner of image as origin (0,0).', + ) + + +class DynamicMask(BaseModel): + mask: Optional[AnyUrl] = Field( + None, + description='Dynamic Brush Application Area (Mask image created by users using the motion brush). The aspect ratio must match the input image.', + ) + trajectories: Optional[List[Trajectory]] = None + + +class TaskInfo(BaseModel): + external_task_id: Optional[str] = None + + class KlingImageGenAspectRatio(str, Enum): field_16_9 = '16:9' field_9_16 = '9:16' @@ -712,278 +930,42 @@ class KlingImageGenModelName(str, Enum): kling_v2 = 'kling-v2' +class KlingImageGenerationsRequest(BaseModel): + aspect_ratio: Optional[KlingImageGenAspectRatio] = '16:9' + callback_url: Optional[AnyUrl] = Field( + None, description='The callback notification address' + ) + human_fidelity: Optional[float] = Field( + 0.45, description='Subject reference similarity', ge=0.0, le=1.0 + ) + image: Optional[str] = Field( + None, description='Reference Image - Base64 encoded string or image URL' + ) + image_fidelity: Optional[float] = Field( + 0.5, description='Reference intensity for user-uploaded images', ge=0.0, le=1.0 + ) + image_reference: Optional[KlingImageGenImageReferenceType] = None + model_name: Optional[KlingImageGenModelName] = 'kling-v1' + n: Optional[int] = Field(1, description='Number of generated images', ge=1, le=9) + negative_prompt: Optional[str] = Field( + None, description='Negative text prompt', max_length=200 + ) + prompt: str = Field(..., description='Positive text prompt', max_length=500) + + class KlingImageResult(BaseModel): index: Optional[int] = Field(None, description='Image Number (0-9)') url: Optional[AnyUrl] = Field(None, description='URL for generated image') -class KlingVirtualTryOnModelName(str, Enum): - kolors_virtual_try_on_v1 = 'kolors-virtual-try-on-v1' - kolors_virtual_try_on_v1_5 = 'kolors-virtual-try-on-v1-5' +class KlingLipSyncMode(str, Enum): + text2video = 'text2video' + audio2video = 'audio2video' -class TaskInfo(BaseModel): - external_task_id: Optional[str] = None - - -class TaskResult(BaseModel): - videos: Optional[List[KlingVideoResult]] = None - - -class Data(BaseModel): - task_id: Optional[str] = Field(None, description='Task ID') - task_status: Optional[KlingTaskStatus] = None - task_info: Optional[TaskInfo] = None - created_at: Optional[int] = Field(None, description='Task creation time') - updated_at: Optional[int] = Field(None, description='Task update time') - task_result: Optional[TaskResult] = None - - -class KlingText2VideoResponse(BaseModel): - code: Optional[int] = Field(None, description='Error code') - message: Optional[str] = Field(None, description='Error message') - request_id: Optional[str] = Field(None, description='Request ID') - data: Optional[Data] = None - - -class Trajectory(BaseModel): - x: Optional[int] = Field( - None, - description='The horizontal coordinate of trajectory point. Based on bottom-left corner of image as origin (0,0).', - ) - y: Optional[int] = Field( - None, - description='The vertical coordinate of trajectory point. Based on bottom-left corner of image as origin (0,0).', - ) - - -class DynamicMask(BaseModel): - mask: Optional[AnyUrl] = Field( - None, - description='Dynamic Brush Application Area (Mask image created by users using the motion brush). The aspect ratio must match the input image.', - ) - trajectories: Optional[List[Trajectory]] = None - - -class Data1(BaseModel): - task_id: Optional[str] = Field(None, description='Task ID') - task_status: Optional[KlingTaskStatus] = None - task_info: Optional[TaskInfo] = None - created_at: Optional[int] = Field(None, description='Task creation time') - updated_at: Optional[int] = Field(None, description='Task update time') - task_result: Optional[TaskResult] = None - - -class KlingImage2VideoResponse(BaseModel): - code: Optional[int] = Field(None, description='Error code') - message: Optional[str] = Field(None, description='Error message') - request_id: Optional[str] = Field(None, description='Request ID') - data: Optional[Data1] = None - - -class KlingVideoExtendRequest(BaseModel): - video_id: Optional[str] = Field( - None, - description='The ID of the video to be extended. Supports videos generated by text-to-video, image-to-video, and previous video extension operations. Cannot exceed 3 minutes total duration after extension.', - ) - prompt: Optional[str] = Field( - None, - description='Positive text prompt for guiding the video extension', - max_length=2500, - ) - negative_prompt: Optional[str] = Field( - None, - description='Negative text prompt for elements to avoid in the extended video', - max_length=2500, - ) - cfg_scale: Optional[KlingVideoGenCfgScale] = Field( - default_factory=lambda: KlingVideoGenCfgScale.model_validate(0.5) - ) - callback_url: Optional[AnyUrl] = Field( - None, - description='The callback notification address. Server will notify when the task status changes.', - ) - - -class Data2(BaseModel): - task_id: Optional[str] = Field(None, description='Task ID') - task_status: Optional[KlingTaskStatus] = None - task_info: Optional[TaskInfo] = None - created_at: Optional[int] = Field(None, description='Task creation time') - updated_at: Optional[int] = Field(None, description='Task update time') - task_result: Optional[TaskResult] = None - - -class KlingVideoExtendResponse(BaseModel): - code: Optional[int] = Field(None, description='Error code') - message: Optional[str] = Field(None, description='Error message') - request_id: Optional[str] = Field(None, description='Request ID') - data: Optional[Data2] = None - - -class KlingLipSyncInputObject(BaseModel): - video_id: Optional[str] = Field( - None, - description='The ID of the video generated by Kling AI. Only supports 5-second and 10-second videos generated within the last 30 days.', - ) - video_url: Optional[str] = Field( - None, - description='Get link for uploaded video. Video files support .mp4/.mov, file size does not exceed 100MB, video length between 2-10s.', - ) - mode: KlingLipSyncMode - text: Optional[str] = Field( - None, - description='Text Content for Lip-Sync Video Generation. Required when mode is text2video. Maximum length is 120 characters.', - ) - voice_id: Optional[str] = Field( - None, - description='Voice ID. Required when mode is text2video. The system offers a variety of voice options to choose from.', - ) - voice_language: Optional[KlingLipSyncVoiceLanguage] = 'en' - voice_speed: Optional[float] = Field( - 1, - description='Speech Rate. Valid range: 0.8~2.0, accurate to one decimal place.', - ge=0.8, - le=2.0, - ) - audio_type: Optional[KlingAudioUploadType] = None - audio_file: Optional[str] = Field( - None, - description='Local Path of Audio File. Supported formats: .mp3/.wav/.m4a/.aac, maximum file size of 5MB. Base64 code.', - ) - audio_url: Optional[str] = Field( - None, - description='Audio File Download URL. Supported formats: .mp3/.wav/.m4a/.aac, maximum file size of 5MB.', - ) - - -class KlingLipSyncRequest(BaseModel): - input: KlingLipSyncInputObject - callback_url: Optional[AnyUrl] = Field( - None, - description='The callback notification address. Server will notify when the task status changes.', - ) - - -class Data3(BaseModel): - task_id: Optional[str] = Field(None, description='Task ID') - task_status: Optional[KlingTaskStatus] = None - task_info: Optional[TaskInfo] = None - created_at: Optional[int] = Field(None, description='Task creation time') - updated_at: Optional[int] = Field(None, description='Task update time') - task_result: Optional[TaskResult] = None - - -class KlingLipSyncResponse(BaseModel): - code: Optional[int] = Field(None, description='Error code') - message: Optional[str] = Field(None, description='Error message') - request_id: Optional[str] = Field(None, description='Request ID') - data: Optional[Data3] = None - - -class KlingSingleImageEffectInput(BaseModel): - model_name: KlingSingleImageEffectModelName - image: str = Field( - ..., - description='Reference Image. URL or Base64 encoded string (without data:image prefix). File size cannot exceed 10MB, resolution not less than 300*300px, aspect ratio between 1:2.5 ~ 2.5:1.', - ) - duration: KlingSingleImageEffectDuration - - -class KlingDualCharacterEffectInput(BaseModel): - model_name: Optional[KlingCharacterEffectModelName] = 'kling-v1' - mode: Optional[KlingVideoGenMode] = 'std' - images: KlingDualCharacterImages - duration: KlingVideoGenDuration - - -class Data4(BaseModel): - task_id: Optional[str] = Field(None, description='Task ID') - task_status: Optional[KlingTaskStatus] = None - task_info: Optional[TaskInfo] = None - created_at: Optional[int] = Field(None, description='Task creation time') - updated_at: Optional[int] = Field(None, description='Task update time') - task_result: Optional[TaskResult] = None - - -class KlingVideoEffectsResponse(BaseModel): - code: Optional[int] = Field(None, description='Error code') - message: Optional[str] = Field(None, description='Error message') - request_id: Optional[str] = Field(None, description='Request ID') - data: Optional[Data4] = None - - -class KlingImageGenerationsRequest(BaseModel): - model_name: Optional[KlingImageGenModelName] = 'kling-v1' - prompt: str = Field(..., description='Positive text prompt', max_length=500) - negative_prompt: Optional[str] = Field( - None, description='Negative text prompt', max_length=200 - ) - image: Optional[str] = Field( - None, description='Reference Image - Base64 encoded string or image URL' - ) - image_reference: Optional[KlingImageGenImageReferenceType] = None - image_fidelity: Optional[float] = Field( - 0.5, description='Reference intensity for user-uploaded images', ge=0.0, le=1.0 - ) - human_fidelity: Optional[float] = Field( - 0.45, description='Subject reference similarity', ge=0.0, le=1.0 - ) - n: Optional[int] = Field(1, description='Number of generated images', ge=1, le=9) - aspect_ratio: Optional[KlingImageGenAspectRatio] = '16:9' - callback_url: Optional[AnyUrl] = Field( - None, description='The callback notification address' - ) - - -class TaskResult5(BaseModel): - images: Optional[List[KlingImageResult]] = None - - -class Data5(BaseModel): - task_id: Optional[str] = Field(None, description='Task ID') - task_status: Optional[KlingTaskStatus] = None - task_status_msg: Optional[str] = Field(None, description='Task status information') - created_at: Optional[int] = Field(None, description='Task creation time') - updated_at: Optional[int] = Field(None, description='Task update time') - task_result: Optional[TaskResult5] = None - - -class KlingImageGenerationsResponse(BaseModel): - code: Optional[int] = Field(None, description='Error code') - message: Optional[str] = Field(None, description='Error message') - request_id: Optional[str] = Field(None, description='Request ID') - data: Optional[Data5] = None - - -class KlingVirtualTryOnRequest(BaseModel): - model_name: Optional[KlingVirtualTryOnModelName] = 'kolors-virtual-try-on-v1' - human_image: str = Field( - ..., description='Reference human image - Base64 encoded string or image URL' - ) - cloth_image: Optional[str] = Field( - None, - description='Reference clothing image - Base64 encoded string or image URL', - ) - callback_url: Optional[AnyUrl] = Field( - None, description='The callback notification address' - ) - - -class Data6(BaseModel): - task_id: Optional[str] = Field(None, description='Task ID') - task_status: Optional[KlingTaskStatus] = None - task_status_msg: Optional[str] = Field(None, description='Task status information') - created_at: Optional[int] = Field(None, description='Task creation time') - updated_at: Optional[int] = Field(None, description='Task update time') - task_result: Optional[TaskResult5] = None - - -class KlingVirtualTryOnResponse(BaseModel): - code: Optional[int] = Field(None, description='Error code') - message: Optional[str] = Field(None, description='Error message') - request_id: Optional[str] = Field(None, description='Request ID') - data: Optional[Data6] = None +class KlingLipSyncVoiceLanguage(str, Enum): + zh = 'zh' + en = 'en' class Config(BaseModel): @@ -1709,29 +1691,29 @@ class Status(str, Enum): class ResourcePackSubscribeInfo(BaseModel): - resource_pack_name: Optional[str] = Field(None, description='Resource package name') - resource_pack_id: Optional[str] = Field(None, description='Resource package ID') - resource_pack_type: Optional[ResourcePackType] = Field( - None, - description='Resource package type (decreasing_total=decreasing total, constant_period=constant periodicity)', - ) - total_quantity: Optional[float] = Field(None, description='Total quantity') - remaining_quantity: Optional[float] = Field( - None, description='Remaining quantity (updated with a 12-hour delay)' - ) - purchase_time: Optional[int] = Field( - None, description='Purchase time, Unix timestamp in ms' - ) effective_time: Optional[int] = Field( None, description='Effective time, Unix timestamp in ms' ) invalid_time: Optional[int] = Field( None, description='Expiration time, Unix timestamp in ms' ) - status: Optional[Status] = Field(None, description='Resource Package Status') + purchase_time: Optional[int] = Field( + None, description='Purchase time, Unix timestamp in ms' + ) + remaining_quantity: Optional[float] = Field( + None, description='Remaining quantity (updated with a 12-hour delay)' + ) + resource_pack_id: Optional[str] = Field(None, description='Resource package ID') + resource_pack_name: Optional[str] = Field(None, description='Resource package name') + resource_pack_type: Optional[ResourcePackType] = Field( + None, + description='Resource package type (decreasing_total=decreasing total, constant_period=constant periodicity)', + ) + status: Optional[Status4] = Field(None, description='Resource Package Status') + total_quantity: Optional[float] = Field(None, description='Total quantity') -class Data7(BaseModel): +class Data3(BaseModel): code: Optional[int] = Field(None, description='Error code; 0 indicates success') msg: Optional[str] = Field(None, description='Error information') resource_pack_subscribe_infos: Optional[List[ResourcePackSubscribeInfo]] = Field( @@ -1741,37 +1723,158 @@ class Data7(BaseModel): class KlingResourcePackageResponse(BaseModel): code: Optional[int] = Field(None, description='Error code; 0 indicates success') + data: Optional[Data3] = None message: Optional[str] = Field(None, description='Error information') request_id: Optional[str] = Field( None, description='Request ID, generated by the system, used to track requests and troubleshoot problems', ) + + +class KlingSingleImageEffectDuration(str, Enum): + field_5 = '5' + + +class KlingSingleImageEffectModelName(str, Enum): + kling_v1_6 = 'kling-v1-6' + + +class KlingSingleImageEffectsScene(str, Enum): + bloombloom = 'bloombloom' + dizzydizzy = 'dizzydizzy' + fuzzyfuzzy = 'fuzzyfuzzy' + squish = 'squish' + expansion = 'expansion' + + +class KlingTaskStatus(str, Enum): + submitted = 'submitted' + processing = 'processing' + succeed = 'succeed' + failed = 'failed' + + +class KlingTextToVideoModelName(str, Enum): + kling_v1 = 'kling-v1' + kling_v1_6 = 'kling-v1-6' + + +class KlingVideoGenAspectRatio(str, Enum): + field_16_9 = '16:9' + field_9_16 = '9:16' + field_1_1 = '1:1' + + +class KlingVideoGenCfgScale(RootModel[float]): + root: float = Field( + ..., + description="Flexibility in video generation. The higher the value, the lower the model's degree of flexibility, and the stronger the relevance to the user's prompt.", + ge=0.0, + le=1.0, + ) + + +class KlingVideoGenDuration(str, Enum): + field_5 = '5' + field_10 = '10' + + +class KlingVideoGenMode(str, Enum): + std = 'std' + pro = 'pro' + + +class KlingVideoGenModelName(str, Enum): + kling_v1 = 'kling-v1' + kling_v1_5 = 'kling-v1-5' + kling_v1_6 = 'kling-v1-6' + kling_v2_master = 'kling-v2-master' + + +class KlingVideoResult(BaseModel): + duration: Optional[str] = Field(None, description='Total video duration') + id: Optional[str] = Field(None, description='Generated video ID') + url: Optional[AnyUrl] = Field(None, description='URL for generated video') + + +class KlingVirtualTryOnModelName(str, Enum): + kolors_virtual_try_on_v1 = 'kolors-virtual-try-on-v1' + kolors_virtual_try_on_v1_5 = 'kolors-virtual-try-on-v1-5' + + +class KlingVirtualTryOnRequest(BaseModel): + callback_url: Optional[AnyUrl] = Field( + None, description='The callback notification address' + ) + cloth_image: Optional[str] = Field( + None, + description='Reference clothing image - Base64 encoded string or image URL', + ) + human_image: str = Field( + ..., description='Reference human image - Base64 encoded string or image URL' + ) + model_name: Optional[KlingVirtualTryOnModelName] = 'kolors-virtual-try-on-v1' + + +class TaskResult6(BaseModel): + images: Optional[List[KlingImageResult]] = None + + +class Data7(BaseModel): + created_at: Optional[int] = Field(None, description='Task creation time') + task_id: Optional[str] = Field(None, description='Task ID') + task_result: Optional[TaskResult6] = None + task_status: Optional[KlingTaskStatus] = None + task_status_msg: Optional[str] = Field(None, description='Task status information') + updated_at: Optional[int] = Field(None, description='Task update time') + + +class KlingVirtualTryOnResponse(BaseModel): + code: Optional[int] = Field(None, description='Error code') data: Optional[Data7] = None + message: Optional[str] = Field(None, description='Error message') + request_id: Optional[str] = Field(None, description='Request ID') -class Object(str, Enum): - event = 'event' +class LumaAspectRatio(str, Enum): + field_1_1 = '1:1' + field_16_9 = '16:9' + field_9_16 = '9:16' + field_4_3 = '4:3' + field_3_4 = '3:4' + field_21_9 = '21:9' + field_9_21 = '9:21' -class Type(str, Enum): - payment_intent_succeeded = 'payment_intent.succeeded' +class LumaAssets(BaseModel): + image: Optional[AnyUrl] = Field(None, description='The URL of the image') + progress_video: Optional[AnyUrl] = Field( + None, description='The URL of the progress video' + ) + video: Optional[AnyUrl] = Field(None, description='The URL of the video') -class StripeRequestInfo(BaseModel): - id: Optional[str] = None - idempotency_key: Optional[str] = None +class GenerationType(str, Enum): + add_audio = 'add_audio' -class Object1(str, Enum): - payment_intent = 'payment_intent' +class LumaAudioGenerationRequest(BaseModel): + callback_url: Optional[AnyUrl] = Field( + None, description='The callback URL for the audio' + ) + generation_type: Optional[GenerationType] = 'add_audio' + negative_prompt: Optional[str] = Field( + None, description='The negative prompt of the audio' + ) + prompt: Optional[str] = Field(None, description='The prompt of the audio') -class StripeAmountDetails(BaseModel): - tip: Optional[Dict[str, Any]] = None +class LumaError(BaseModel): + detail: Optional[str] = Field(None, description='The error message') -class Object2(str, Enum): - charge = 'charge' +class Type11(str, Enum): + generation = 'generation' class LumaImageIdentity(BaseModel): @@ -2042,9 +2145,9 @@ class OpenAIImageEditRequest(BaseModel): ) -class MinimaxFileRetrieveResponse(BaseModel): - file: File - base_resp: MinimaxBaseResponse +class Moderation(str, Enum): + low = 'low' + auto = 'auto' class Status1(str, Enum): @@ -2543,10 +2646,60 @@ class RecraftTransformModel(str, Enum): class RecraftImageFormat(str, Enum): webp = 'webp' - png = 'png' + jpeg = 'jpeg' -class RecraftResponseFormat(str, Enum): +class OpenAIImageEditRequest(BaseModel): + background: Optional[str] = Field( + None, description='Background transparency', examples=['opaque'] + ) + model: str = Field( + ..., description='The model to use for image editing', examples=['gpt-image-1'] + ) + moderation: Optional[Moderation] = Field( + None, description='Content moderation setting', examples=['auto'] + ) + n: Optional[int] = Field( + None, description='The number of images to generate', examples=[1] + ) + output_compression: Optional[int] = Field( + None, description='Compression level for JPEG or WebP (0-100)', examples=[100] + ) + output_format: Optional[OutputFormat1] = Field( + None, description='Format of the output image', examples=['png'] + ) + prompt: str = Field( + ..., + description='A text description of the desired edit', + examples=['Give the rocketship rainbow coloring'], + ) + quality: Optional[str] = Field( + None, description='The quality of the edited image', examples=['low'] + ) + size: Optional[str] = Field( + None, description='Size of the output image', examples=['1024x1024'] + ) + user: Optional[str] = Field( + None, + description='A unique identifier for end-user monitoring', + examples=['user-1234'], + ) + + +class Background(str, Enum): + transparent = 'transparent' + opaque = 'opaque' + + +class Quality(str, Enum): + low = 'low' + medium = 'medium' + high = 'high' + standard = 'standard' + hd = 'hd' + + +class ResponseFormat(str, Enum): url = 'url' b64_json = 'b64_json' @@ -2863,11 +3016,8 @@ class Quality(str, Enum): class Style(str, Enum): - anime = 'anime' - field_3d_animation = '3d_animation' - clay = 'clay' - comic = 'comic' - cyberpunk = 'cyberpunk' + vivid = 'vivid' + natural = 'natural' class PixverseImageVideoRequest(BaseModel): @@ -2973,7 +3123,8 @@ class Resp2(BaseModel): size: Optional[int] = None status: Optional[Status2] = Field( None, - description='Video generation status codes:\n* 1 - Generation successful\n* 5 - Generating\n* 6 - Deleted\n* 7 - Contents moderation failed\n* 8 - Generation failed\n', + description='The number of images to generate (1-10). Only 1 supported for dall-e-3.', + examples=[1], ) style: Optional[str] = None url: Optional[str] = None @@ -3060,18 +3211,10 @@ class Veo2GenVidPollRequest(BaseModel): 'projects/PROJECT_ID/locations/us-central1/publishers/google/models/MODEL_ID/operations/OPERATION_ID' ], ) - - -class Video(BaseModel): - gcsUri: Optional[str] = Field(None, description='Cloud Storage URI of the video') - bytesBase64Encoded: Optional[str] = Field( - None, description='Base64-encoded video content' + response_format: Optional[ResponseFormat] = Field( + None, description='Response format of image data', examples=['b64_json'] ) - mimeType: Optional[str] = Field(None, description='Video MIME type') - - -class Response(BaseModel): - field_type: Optional[str] = Field( + size: Optional[str] = Field( None, description='The callback URL of the generation, a POST request with Generation object will be sent to the callback URL when the generation is dreaming, completed, or failed', ) @@ -3163,13 +3306,13 @@ class RunwayPromptImageObject( class Datum3(BaseModel): b64_json: Optional[str] = Field(None, description='Base64 encoded image data') - url: Optional[str] = Field(None, description='URL of the image') revised_prompt: Optional[str] = Field(None, description='Revised prompt') + url: Optional[str] = Field(None, description='URL of the image') class InputTokensDetails(BaseModel): - text_tokens: Optional[int] = None image_tokens: Optional[int] = None + text_tokens: Optional[int] = None class Usage(BaseModel): @@ -3180,143 +3323,204 @@ class Usage(BaseModel): class OpenAIImageGenerationResponse(BaseModel): - data: Optional[List[Datum3]] = None + data: Optional[List[Datum2]] = None usage: Optional[Usage] = None -class Quality3(str, Enum): - low = 'low' - medium = 'medium' - high = 'high' - standard = 'standard' - hd = 'hd' +class OpenAIModels(str, Enum): + gpt_4 = 'gpt-4' + gpt_4_0314 = 'gpt-4-0314' + gpt_4_0613 = 'gpt-4-0613' + gpt_4_32k = 'gpt-4-32k' + gpt_4_32k_0314 = 'gpt-4-32k-0314' + gpt_4_32k_0613 = 'gpt-4-32k-0613' + gpt_4_0125_preview = 'gpt-4-0125-preview' + gpt_4_turbo = 'gpt-4-turbo' + gpt_4_turbo_2024_04_09 = 'gpt-4-turbo-2024-04-09' + gpt_4_turbo_preview = 'gpt-4-turbo-preview' + gpt_4_1106_preview = 'gpt-4-1106-preview' + gpt_4_vision_preview = 'gpt-4-vision-preview' + gpt_3_5_turbo = 'gpt-3.5-turbo' + gpt_3_5_turbo_16k = 'gpt-3.5-turbo-16k' + gpt_3_5_turbo_0301 = 'gpt-3.5-turbo-0301' + gpt_3_5_turbo_0613 = 'gpt-3.5-turbo-0613' + gpt_3_5_turbo_1106 = 'gpt-3.5-turbo-1106' + gpt_3_5_turbo_0125 = 'gpt-3.5-turbo-0125' + gpt_3_5_turbo_16k_0613 = 'gpt-3.5-turbo-16k-0613' + gpt_4_1 = 'gpt-4.1' + gpt_4_1_mini = 'gpt-4.1-mini' + gpt_4_1_nano = 'gpt-4.1-nano' + gpt_4_1_2025_04_14 = 'gpt-4.1-2025-04-14' + gpt_4_1_mini_2025_04_14 = 'gpt-4.1-mini-2025-04-14' + gpt_4_1_nano_2025_04_14 = 'gpt-4.1-nano-2025-04-14' + o1 = 'o1' + o1_mini = 'o1-mini' + o1_preview = 'o1-preview' + o1_pro = 'o1-pro' + o1_2024_12_17 = 'o1-2024-12-17' + o1_preview_2024_09_12 = 'o1-preview-2024-09-12' + o1_mini_2024_09_12 = 'o1-mini-2024-09-12' + o1_pro_2025_03_19 = 'o1-pro-2025-03-19' + o3 = 'o3' + o3_mini = 'o3-mini' + o3_2025_04_16 = 'o3-2025-04-16' + o3_mini_2025_01_31 = 'o3-mini-2025-01-31' + o4_mini = 'o4-mini' + o4_mini_2025_04_16 = 'o4-mini-2025-04-16' + gpt_4o = 'gpt-4o' + gpt_4o_mini = 'gpt-4o-mini' + gpt_4o_2024_11_20 = 'gpt-4o-2024-11-20' + gpt_4o_2024_08_06 = 'gpt-4o-2024-08-06' + gpt_4o_2024_05_13 = 'gpt-4o-2024-05-13' + gpt_4o_mini_2024_07_18 = 'gpt-4o-mini-2024-07-18' + gpt_4o_audio_preview = 'gpt-4o-audio-preview' + gpt_4o_audio_preview_2024_10_01 = 'gpt-4o-audio-preview-2024-10-01' + gpt_4o_audio_preview_2024_12_17 = 'gpt-4o-audio-preview-2024-12-17' + gpt_4o_mini_audio_preview = 'gpt-4o-mini-audio-preview' + gpt_4o_mini_audio_preview_2024_12_17 = 'gpt-4o-mini-audio-preview-2024-12-17' + gpt_4o_search_preview = 'gpt-4o-search-preview' + gpt_4o_mini_search_preview = 'gpt-4o-mini-search-preview' + gpt_4o_search_preview_2025_03_11 = 'gpt-4o-search-preview-2025-03-11' + gpt_4o_mini_search_preview_2025_03_11 = 'gpt-4o-mini-search-preview-2025-03-11' + computer_use_preview = 'computer-use-preview' + computer_use_preview_2025_03_11 = 'computer-use-preview-2025-03-11' + chatgpt_4o_latest = 'chatgpt-4o-latest' -class OutputFormat1(str, Enum): - png = 'png' - webp = 'webp' - jpeg = 'jpeg' +class Reason(str, Enum): + max_output_tokens = 'max_output_tokens' + content_filter = 'content_filter' -class Moderation(str, Enum): - low = 'low' - auto = 'auto' - - -class Background(str, Enum): - transparent = 'transparent' - opaque = 'opaque' - - -class ResponseFormat(str, Enum): - url = 'url' - b64_json = 'b64_json' - - -class Style3(str, Enum): - vivid = 'vivid' - natural = 'natural' - - -class OpenAIImageGenerationRequest(BaseModel): - model: Optional[str] = Field( - None, description='The model to use for image generation', examples=['dall-e-3'] +class IncompleteDetails(BaseModel): + reason: Optional[Reason] = Field( + None, description='The reason why the response is incomplete.' ) - prompt: str = Field( + + +class Object(str, Enum): + response = 'response' + + +class Status6(str, Enum): + completed = 'completed' + failed = 'failed' + in_progress = 'in_progress' + incomplete = 'incomplete' + + +class Type13(str, Enum): + output_audio = 'output_audio' + + +class OutputAudioContent(BaseModel): + data: str = Field(..., description='Base64-encoded audio data') + transcript: str = Field(..., description='Transcript of the audio') + type: Type13 = Field(..., description='The type of output content') + + +class Role2(str, Enum): + assistant = 'assistant' + + +class Type14(str, Enum): + message = 'message' + + +class Type15(str, Enum): + output_text = 'output_text' + + +class OutputTextContent(BaseModel): + text: str = Field(..., description='The text content') + type: Type15 = Field(..., description='The type of output content') + + +class AspectRatio1(RootModel[float]): + root: float = Field( ..., - description='A text description of the desired image', - examples=['Draw a rocket in front of a blackhole in deep space'], - ) - n: Optional[int] = Field( - None, - description='The number of images to generate (1-10). Only 1 supported for dall-e-3.', - examples=[1], - ) - quality: Optional[Quality3] = Field( - None, description='The quality of the generated image', examples=['high'] - ) - size: Optional[str] = Field( - None, - description='Size of the image (e.g., 1024x1024, 1536x1024, auto)', - examples=['1024x1536'], - ) - output_format: Optional[OutputFormat1] = Field( - None, description='Format of the output image', examples=['png'] - ) - output_compression: Optional[int] = Field( - None, description='Compression level for JPEG or WebP (0-100)', examples=[100] - ) - moderation: Optional[Moderation] = Field( - None, description='Content moderation setting', examples=['auto'] - ) - background: Optional[Background] = Field( - None, description='Background transparency', examples=['opaque'] - ) - response_format: Optional[ResponseFormat] = Field( - None, description='Response format of image data', examples=['b64_json'] - ) - style: Optional[Style3] = Field( - None, description='Style of the image (only for dall-e-3)', examples=['vivid'] - ) - user: Optional[str] = Field( - None, - description='A unique identifier for end-user monitoring', - examples=['user-1234'], + description='Aspect ratio (width / height)', + ge=0.4, + le=2.5, + title='Aspectratio', ) -class OpenAIImageEditRequest(BaseModel): - model: str = Field( - ..., description='The model to use for image editing', examples=['gpt-image-1'] - ) - prompt: str = Field( - ..., - description='A text description of the desired edit', - examples=['Give the rocketship rainbow coloring'], - ) - n: Optional[int] = Field( - None, description='The number of images to generate', examples=[1] - ) - quality: Optional[str] = Field( - None, description='The quality of the edited image', examples=['low'] - ) - size: Optional[str] = Field( - None, description='Size of the output image', examples=['1024x1024'] - ) - output_format: Optional[OutputFormat1] = Field( - None, description='Format of the output image', examples=['png'] - ) - output_compression: Optional[int] = Field( - None, description='Compression level for JPEG or WebP (0-100)', examples=[100] - ) - moderation: Optional[Moderation] = Field( - None, description='Content moderation setting', examples=['auto'] - ) - background: Optional[str] = Field( - None, description='Background transparency', examples=['opaque'] - ) - user: Optional[str] = Field( - None, - description='A unique identifier for end-user monitoring', - examples=['user-1234'], - ) +class IngredientsMode(str, Enum): + creative = 'creative' + precise = 'precise' -class CustomerStorageResourceResponse(BaseModel): - download_url: Optional[str] = Field( +class PikaBodyGenerate22C2vGenerate22PikascenesPost(BaseModel): + aspectRatio: Optional[AspectRatio1] = Field( + None, description='Aspect ratio (width / height)', title='Aspectratio' + ) + duration: Optional[int] = Field(5, title='Duration') + images: Optional[List[StrictBytes]] = Field(None, title='Images') + ingredientsMode: IngredientsMode = Field(..., title='Ingredientsmode') + negativePrompt: Optional[str] = Field(None, title='Negativeprompt') + promptText: Optional[str] = Field(None, title='Prompttext') + resolution: Optional[str] = Field('1080p', title='Resolution') + seed: Optional[int] = Field(None, title='Seed') + + +class PikaBodyGeneratePikadditionsGeneratePikadditionsPost(BaseModel): + image: Optional[StrictBytes] = Field(None, title='Image') + negativePrompt: Optional[str] = Field(None, title='Negativeprompt') + promptText: Optional[str] = Field(None, title='Prompttext') + seed: Optional[int] = Field(None, title='Seed') + video: Optional[StrictBytes] = Field(None, title='Video') + + +class PikaBodyGeneratePikaswapsGeneratePikaswapsPost(BaseModel): + image: Optional[StrictBytes] = Field(None, title='Image') + modifyRegionMask: Optional[StrictBytes] = Field( None, - description='The signed URL to use for downloading the file from the specified path', + description='A mask image that specifies the region to modify, where the mask is white and the background is black', + title='Modifyregionmask', ) - upload_url: Optional[str] = Field( + modifyRegionRoi: Optional[str] = Field( None, - description='The signed URL to use for uploading the file to the specified path', - ) - expires_at: Optional[datetime] = Field( - None, description='When the signed URL will expire' - ) - existing_file: Optional[bool] = Field( - None, description='Whether an existing file with the same hash was found' + description='Plaintext description of the object / region to modify', + title='Modifyregionroi', ) + negativePrompt: Optional[str] = Field(None, title='Negativeprompt') + promptText: Optional[str] = Field(None, title='Prompttext') + seed: Optional[int] = Field(None, title='Seed') + video: Optional[StrictBytes] = Field(None, title='Video') + + +class PikaDurationEnum(int, Enum): + integer_5 = 5 + integer_10 = 10 + + +class PikaGenerateResponse(BaseModel): + video_id: str = Field(..., title='Video Id') + + +class PikaResolutionEnum(str, Enum): + field_1080p = '1080p' + field_720p = '720p' + + +class PikaStatusEnum(str, Enum): + queued = 'queued' + started = 'started' + finished = 'finished' + + +class PikaValidationError(BaseModel): + loc: List[Union[str, int]] = Field(..., title='Location') + msg: str = Field(..., title='Message') + type: str = Field(..., title='Error Type') + + +class PikaVideoResponse(BaseModel): + id: str = Field(..., title='Id') + progress: Optional[int] = Field(None, title='Progress') + status: PikaStatusEnum + url: Optional[str] = Field(None, title='Url') class Pikaffect(str, Enum): @@ -3338,92 +3542,135 @@ class Pikaffect(str, Enum): Tear = 'Tear' -class PikaBodyGeneratePikaffectsGeneratePikaffectsPost(BaseModel): - image: Optional[StrictBytes] = Field(None, title='Image') - pikaffect: Optional[Pikaffect] = Field(None, title='Pikaffect') - promptText: Optional[str] = Field(None, title='Prompttext') - negativePrompt: Optional[str] = Field(None, title='Negativeprompt') - seed: Optional[int] = Field(None, title='Seed') +class Resp(BaseModel): + img_id: Optional[int] = None -class PikaGenerateResponse(BaseModel): - video_id: str = Field(..., title='Video Id') +class PixverseImageUploadResponse(BaseModel): + ErrCode: Optional[int] = None + ErrMsg: Optional[str] = None + Resp_1: Optional[Resp] = Field(None, alias='Resp') -class PikaBodyGeneratePikadditionsGeneratePikadditionsPost(BaseModel): - video: Optional[StrictBytes] = Field(None, title='Video') - image: Optional[StrictBytes] = Field(None, title='Image') - promptText: Optional[str] = Field(None, title='Prompttext') - negativePrompt: Optional[str] = Field(None, title='Negativeprompt') - seed: Optional[int] = Field(None, title='Seed') - - -class PikaBodyGeneratePikaswapsGeneratePikaswapsPost(BaseModel): - video: Optional[StrictBytes] = Field(None, title='Video') - image: Optional[StrictBytes] = Field(None, title='Image') - promptText: Optional[str] = Field(None, title='Prompttext') - modifyRegionMask: Optional[StrictBytes] = Field( - None, - description='A mask image that specifies the region to modify, where the mask is white and the background is black', - title='Modifyregionmask', - ) - modifyRegionRoi: Optional[str] = Field( - None, - description='Plaintext description of the object / region to modify', - title='Modifyregionroi', - ) - negativePrompt: Optional[str] = Field(None, title='Negativeprompt') - seed: Optional[int] = Field(None, title='Seed') - - -class IngredientsMode(str, Enum): - creative = 'creative' - precise = 'precise' - - -class AspectRatio1(RootModel[float]): - root: float = Field( - ..., - description='Aspect ratio (width / height)', - ge=0.4, - le=2.5, - title='Aspectratio', - ) - - -class PikaBodyGenerate22C2vGenerate22PikascenesPost(BaseModel): - images: Optional[List[StrictBytes]] = Field(None, title='Images') - ingredientsMode: IngredientsMode = Field(..., title='Ingredientsmode') - promptText: Optional[str] = Field(None, title='Prompttext') - negativePrompt: Optional[str] = Field(None, title='Negativeprompt') - seed: Optional[int] = Field(None, title='Seed') - resolution: Optional[str] = Field('1080p', title='Resolution') - duration: Optional[int] = Field(5, title='Duration') - aspectRatio: Optional[AspectRatio1] = Field( - None, description='Aspect ratio (width / height)', title='Aspectratio' - ) - - -class PikaStatusEnum(str, Enum): - queued = 'queued' - started = 'started' - finished = 'finished' - - -class PikaValidationError(BaseModel): - loc: List[Union[str, int]] = Field(..., title='Location') - msg: str = Field(..., title='Message') - type: str = Field(..., title='Error Type') - - -class PikaResolutionEnum(str, Enum): - field_1080p = '1080p' - field_720p = '720p' - - -class PikaDurationEnum(int, Enum): +class Duration(int, Enum): integer_5 = 5 - integer_10 = 10 + integer_8 = 8 + + +class Model1(str, Enum): + v3_5 = 'v3.5' + + +class MotionMode(str, Enum): + normal = 'normal' + fast = 'fast' + + +class Quality1(str, Enum): + field_360p = '360p' + field_540p = '540p' + field_720p = '720p' + field_1080p = '1080p' + + +class Style1(str, Enum): + anime = 'anime' + field_3d_animation = '3d_animation' + clay = 'clay' + comic = 'comic' + cyberpunk = 'cyberpunk' + + +class PixverseImageVideoRequest(BaseModel): + duration: Duration + img_id: int + model: Model1 + motion_mode: Optional[MotionMode] = None + prompt: str + quality: Quality1 + seed: Optional[int] = None + style: Optional[Style1] = None + template_id: Optional[int] = None + water_mark: Optional[bool] = None + + +class AspectRatio2(str, Enum): + field_16_9 = '16:9' + field_4_3 = '4:3' + field_1_1 = '1:1' + field_3_4 = '3:4' + field_9_16 = '9:16' + + +class PixverseTextVideoRequest(BaseModel): + aspect_ratio: AspectRatio2 + duration: Duration + model: Model1 + motion_mode: Optional[MotionMode] = None + negative_prompt: Optional[str] = None + prompt: str + quality: Quality1 + seed: Optional[int] = None + style: Optional[Style1] = None + template_id: Optional[int] = None + water_mark: Optional[bool] = None + + +class PixverseTransitionVideoRequest(BaseModel): + duration: Duration + first_frame_img: int + last_frame_img: int + model: Model1 + motion_mode: MotionMode + prompt: str + quality: Quality1 + seed: int + style: Optional[Style1] = None + template_id: Optional[int] = None + water_mark: Optional[bool] = None + + +class Resp1(BaseModel): + video_id: Optional[int] = None + + +class PixverseVideoResponse(BaseModel): + ErrCode: Optional[int] = None + ErrMsg: Optional[str] = None + Resp: Optional[Resp1] = None + + +class Status7(int, Enum): + integer_1 = 1 + integer_5 = 5 + integer_6 = 6 + integer_7 = 7 + integer_8 = 8 + + +class Resp2(BaseModel): + create_time: Optional[str] = None + id: Optional[int] = None + modify_time: Optional[str] = None + negative_prompt: Optional[str] = None + outputHeight: Optional[int] = None + outputWidth: Optional[int] = None + prompt: Optional[str] = None + resolution_ratio: Optional[int] = None + seed: Optional[int] = None + size: Optional[int] = None + status: Optional[Status7] = Field( + None, + description='Video generation status codes:\n* 1 - Generation successful\n* 5 - Generating\n* 6 - Deleted\n* 7 - Contents moderation failed\n* 8 - Generation failed\n', + ) + style: Optional[str] = None + url: Optional[str] = None + + +class PixverseVideoResultResponse(BaseModel): + ErrCode: Optional[int] = None + ErrMsg: Optional[str] = None + Resp: Optional[Resp2] = None class RgbItem(RootModel[int]): @@ -3434,213 +3681,273 @@ class RGBColor(BaseModel): rgb: List[RgbItem] = Field(..., max_length=3, min_length=3) -class StabilityStabilityClientID(RootModel[str]): - root: str = Field( +class GenerateSummary(str, Enum): + auto = 'auto' + concise = 'concise' + detailed = 'detailed' + + +class Summary(str, Enum): + auto = 'auto' + concise = 'concise' + detailed = 'detailed' + + +class ReasoningEffort(str, Enum): + low = 'low' + medium = 'medium' + high = 'high' + + +class Status8(str, Enum): + in_progress = 'in_progress' + completed = 'completed' + incomplete = 'incomplete' + + +class Type16(str, Enum): + summary_text = 'summary_text' + + +class SummaryItem(BaseModel): + text: str = Field( ..., - description='The name of your application, used to help us communicate app-specific debugging or moderation issues to you.', - examples=['my-awesome-app'], - max_length=256, + description='A short summary of the reasoning used by the model when generating\nthe response.\n', + ) + type: Type16 = Field( + ..., description='The type of the object. Always `summary_text`.\n' ) -class StabilityStabilityClientUserID(RootModel[str]): - root: str = Field( - ..., - description='A unique identifier for your end user. Used to help us communicate user-specific debugging or moderation issues to you. Feel free to obfuscate this value to protect user privacy.', - examples=['DiscordUser#9999'], - max_length=256, - ) +class Type17(str, Enum): + reasoning = 'reasoning' -class StabilityStabilityClientVersion(RootModel[str]): - root: str = Field( - ..., - description='The version of your application, used to help us communicate version-specific debugging or moderation issues to you.', - examples=['1.2.1'], - max_length=256, - ) - - -class Name(str, Enum): - content_moderation = 'content_moderation' - - -class StabilityContentModerationResponse(BaseModel): +class ReasoningItem(BaseModel): id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new) you file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, + ..., description='The unique identifier of the reasoning content.\n' ) - name: Name = Field( - ..., - description='Our content moderation system has flagged some part of your request and subsequently denied it. You were not charged for this request. While this may at times be frustrating, it is necessary to maintain the integrity of our platform and ensure a safe experience for all users. If you would like to provide feedback, please use the [Support Form](https://kb.stability.ai/knowledge-base/kb-tickets/new).', + status: Optional[Status8] = Field( + None, + description='The status of the item. One of `in_progress`, `completed`, or\n`incomplete`. Populated when items are returned via API.\n', ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, + summary: List[SummaryItem] = Field(..., description='Reasoning text contents.\n') + type: Type17 = Field( + ..., description='The type of the object. Always `reasoning`.\n' ) +class Controls(BaseModel): + artistic_level: Optional[int] = Field( + None, + description='Defines artistic tone of your image. At a simple level, the person looks straight at the camera in a static and clean style. Dynamic and eccentric levels introduce movement and creativity.', + ge=0, + le=5, + ) + background_color: Optional[RGBColor] = None + colors: Optional[List[RGBColor]] = Field( + None, description='An array of preferable colors' + ) + no_text: Optional[bool] = Field(None, description='Do not embed text layouts') + + +class RecraftImageGenerationRequest(BaseModel): + controls: Optional[Controls] = Field( + None, description='The controls for the generated image' + ) + model: str = Field( + ..., description='The model to use for generation (e.g., "recraftv3")' + ) + n: int = Field(..., description='The number of images to generate', ge=1, le=4) + prompt: str = Field( + ..., description='The text prompt describing the image to generate' + ) + size: str = Field( + ..., description='The size of the generated image (e.g., "1024x1024")' + ) + style: Optional[str] = Field( + None, + description='The style to apply to the generated image (e.g., "digital_illustration")', + ) + style_id: Optional[str] = Field( + None, + description='The style ID to apply to the generated image (e.g., "123e4567-e89b-12d3-a456-426614174000"). If style_id is provided, style should not be provided.', + ) + + +class Datum3(BaseModel): + image_id: Optional[str] = Field( + None, description='Unique identifier for the generated image' + ) + url: Optional[str] = Field(None, description='URL to access the generated image') + + +class RecraftImageGenerationResponse(BaseModel): + created: int = Field( + ..., description='Unix timestamp when the generation was created' + ) + credits: int = Field(..., description='Number of credits used for the generation') + data: List[Datum3] = Field(..., description='Array of generated image information') + + class RenderingSpeed(str, Enum): BALANCED = 'BALANCED' TURBO = 'TURBO' QUALITY = 'QUALITY' -class StabilityCreativity(RootModel[float]): - root: float = Field( +class ResponseErrorCode(str, Enum): + server_error = 'server_error' + rate_limit_exceeded = 'rate_limit_exceeded' + invalid_prompt = 'invalid_prompt' + vector_store_timeout = 'vector_store_timeout' + invalid_image = 'invalid_image' + invalid_image_format = 'invalid_image_format' + invalid_base64_image = 'invalid_base64_image' + invalid_image_url = 'invalid_image_url' + image_too_large = 'image_too_large' + image_too_small = 'image_too_small' + image_parse_error = 'image_parse_error' + image_content_policy_violation = 'image_content_policy_violation' + invalid_image_mode = 'invalid_image_mode' + image_file_too_large = 'image_file_too_large' + unsupported_image_media_type = 'unsupported_image_media_type' + empty_image_file = 'empty_image_file' + failed_to_download_image = 'failed_to_download_image' + image_file_not_found = 'image_file_not_found' + + +class Type18(str, Enum): + json_object = 'json_object' + + +class ResponseFormatJsonObject(BaseModel): + type: Type18 = Field( ..., - description='Controls the likelihood of creating additional details not heavily conditioned by the init image.', - ge=0.2, - le=0.5, + description='The type of response format being defined. Always `json_object`.', ) -class StabilityGenerationID(RootModel[str]): - root: str = Field( - ..., - description='The `id` of a generation, typically used for async generations, that can be used to check the status of the generation or retrieve the result.', - examples=['a6dc6c6e20acda010fe14d71f180658f2896ed9b4ec25aa99a6ff06c796987c4'], - max_length=64, - min_length=64, +class ResponseFormatJsonSchemaSchema(BaseModel): + pass + model_config = ConfigDict( + extra='allow', ) -class Mode(str, Enum): - text_to_image = 'text-to-image' - image_to_image = 'image-to-image' +class Type19(str, Enum): + text = 'text' -class AspectRatio2(str, Enum): - field_21_9 = '21:9' - field_16_9 = '16:9' - field_3_2 = '3:2' - field_5_4 = '5:4' - field_1_1 = '1:1' - field_4_5 = '4:5' - field_2_3 = '2:3' - field_9_16 = '9:16' - field_9_21 = '9:21' +class ResponseFormatText(BaseModel): + type: Type19 = Field( + ..., description='The type of response format being defined. Always `text`.' + ) -class Model4(str, Enum): - sd3_5_large = 'sd3.5-large' - sd3_5_large_turbo = 'sd3.5-large-turbo' - sd3_5_medium = 'sd3.5-medium' +class Truncation1(str, Enum): + auto = 'auto' + disabled = 'disabled' -class OutputFormat3(str, Enum): - png = 'png' - jpeg = 'jpeg' - - -class StylePreset(str, Enum): - enhance = 'enhance' - anime = 'anime' - photographic = 'photographic' - digital_art = 'digital-art' - comic_book = 'comic-book' - fantasy_art = 'fantasy-art' - line_art = 'line-art' - analog_film = 'analog-film' - neon_punk = 'neon-punk' - isometric = 'isometric' - low_poly = 'low-poly' - origami = 'origami' - modeling_compound = 'modeling-compound' - cinematic = 'cinematic' - field_3d_model = '3d-model' - pixel_art = 'pixel-art' - tile_texture = 'tile-texture' - - -class StabilityImageGenrationSD3Request(BaseModel): - prompt: str = Field( +class InputTokensDetails1(BaseModel): + cached_tokens: int = Field( ..., - description='What you wish to see in the output image. A strong, descriptive prompt that clearly defines\nelements, colors, and subjects will lead to better results.', - max_length=10000, + description='The number of tokens that were retrieved from the cache. \n[More on prompt caching](/docs/guides/prompt-caching).\n', + ) + + +class OutputTokensDetails(BaseModel): + reasoning_tokens: int = Field(..., description='The number of reasoning tokens.') + + +class ResponseUsage(BaseModel): + input_tokens: int = Field(..., description='The number of input tokens.') + input_tokens_details: InputTokensDetails1 = Field( + ..., description='A detailed breakdown of the input tokens.' + ) + output_tokens: int = Field(..., description='The number of output tokens.') + output_tokens_details: OutputTokensDetails = Field( + ..., description='A detailed breakdown of the output tokens.' + ) + total_tokens: int = Field(..., description='The total number of tokens used.') + + +class RunwayAspectRatioEnum(str, Enum): + field_1280_720 = '1280:720' + field_720_1280 = '720:1280' + field_1104_832 = '1104:832' + field_832_1104 = '832:1104' + field_960_960 = '960:960' + field_1584_672 = '1584:672' + field_1280_768 = '1280:768' + field_768_1280 = '768:1280' + + +class RunwayDurationEnum(int, Enum): + integer_5 = 5 + integer_10 = 10 + + +class RunwayImageToVideoResponse(BaseModel): + id: Optional[str] = Field(None, description='Task ID') + + +class RunwayModelEnum(str, Enum): + gen4_turbo = 'gen4_turbo' + gen3a_turbo = 'gen3a_turbo' + + +class Position(str, Enum): + first = 'first' + last = 'last' + + +class RunwayPromptImageDetailedObject(BaseModel): + position: Position = Field( + ..., + description="The position of the image in the output video. 'last' is currently supported for gen3a_turbo only.", + ) + uri: str = Field( + ..., description='A HTTPS URL or data URI containing an encoded image.' + ) + + +class RunwayPromptImageObject( + RootModel[Union[str, List[RunwayPromptImageDetailedObject]]] +): + root: Union[str, List[RunwayPromptImageDetailedObject]] = Field( + ..., + description='Image(s) to use for the video generation. Can be a single URI or an array of image objects with positions.', + ) + + +class RunwayTaskStatusEnum(str, Enum): + SUCCEEDED = 'SUCCEEDED' + RUNNING = 'RUNNING' + FAILED = 'FAILED' + PENDING = 'PENDING' + CANCELLED = 'CANCELLED' + THROTTLED = 'THROTTLED' + + +class RunwayTaskStatusResponse(BaseModel): + createdAt: Optional[datetime] = Field(None, description='Task creation timestamp') + id: Optional[str] = Field(None, description='Task ID') + output: Optional[List[str]] = Field(None, description='Array of output video URLs') + status: Optional[RunwayTaskStatusEnum] = None + + +class StabilityError(BaseModel): + errors: List[str] = Field( + ..., + description='One or more error messages indicating what went wrong.', + examples=[[{'some-field': 'is required'}]], min_length=1, ) - mode: Optional[Mode] = Field( - 'text-to-image', - description='Controls whether this is a text-to-image or image-to-image generation, which affects which parameters are required:\n- **text-to-image** requires only the `prompt` parameter\n- **image-to-image** requires the `prompt`, `image`, and `strength` parameters', - title='GenerationMode', - ) - image: Optional[StrictBytes] = Field( - None, - description='The image to use as the starting point for the generation.\n\nSupported formats:\n\n\n\n - jpeg\n - png\n - webp\n\nSupported dimensions:\n\n\n\n - Every side must be at least 64 pixels\n\n> **Important:** This parameter is only valid for **image-to-image** requests.', - ) - strength: Optional[float] = Field( - None, - description='Sometimes referred to as _denoising_, this parameter controls how much influence the\n`image` parameter has on the generated image. A value of 0 would yield an image that\nis identical to the input. A value of 1 would be as if you passed in no image at all.\n\n> **Important:** This parameter is only valid for **image-to-image** requests.', - ge=0.0, - le=1.0, - ) - aspect_ratio: Optional[AspectRatio2] = Field( - '1:1', - description='Controls the aspect ratio of the generated image. Defaults to 1:1.\n\n> **Important:** This parameter is only valid for **text-to-image** requests.', - ) - model: Optional[Model4] = Field( - 'sd3.5-large', - description='The model to use for generation.\n\n- `sd3.5-large` requires 6.5 credits per generation\n- `sd3.5-large-turbo` requires 4 credits per generation\n- `sd3.5-medium` requires 3.5 credits per generation\n- As of the April 17, 2025, `sd3-large`, `sd3-large-turbo` and `sd3-medium`\n\n\n\n are re-routed to their `sd3.5-[model version]` equivalent, at the same price.', - ) - seed: Optional[float] = Field( - 0, - description="A specific value that is used to guide the 'randomness' of the generation. (Omit this parameter or pass `0` to use a random seed.)", - ge=0.0, - le=4294967294.0, - ) - output_format: Optional[OutputFormat3] = Field( - 'png', description='Dictates the `content-type` of the generated image.' - ) - style_preset: Optional[StylePreset] = Field( - None, description='Guides the image model towards a particular style.' - ) - negative_prompt: Optional[str] = Field( - None, - description='Keywords of what you **do not** wish to see in the output image.\nThis is an advanced feature.', - max_length=10000, - ) - cfg_scale: Optional[float] = Field( - None, - description='How strictly the diffusion process adheres to the prompt text (higher values keep your image closer to your prompt). The _Large_ and _Medium_ models use a default of `4`. The _Turbo_ model uses a default of `1`.', - ge=1.0, - le=10.0, - ) - - -class FinishReason(str, Enum): - SUCCESS = 'SUCCESS' - CONTENT_FILTERED = 'CONTENT_FILTERED' - - -class StabilityImageGenrationSD3Response200(BaseModel): - image: str = Field( - ..., - description='The generated image, encoded to base64.', - examples=['AAAAIGZ0eXBpc29tAAACAGlzb21pc28yYXZjMW1...'], - ) - seed: Optional[float] = Field( - 0, - description='The seed used as random noise for this generation.', - examples=[343940597], - ge=0.0, - le=4294967294.0, - ) - finish_reason: FinishReason = Field( - ..., - description='The reason the generation finished.\n\n- `SUCCESS` = successful generation.\n- `CONTENT_FILTERED` = successful generation, however the output violated our content moderation\npolicy and has been blurred as a result.', - examples=['SUCCESS'], - ) - - -class StabilityImageGenrationSD3Response400(BaseModel): id: str = Field( ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', + description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new) you file, as it will greatly assist us in diagnosing the root cause of the problem.\n', examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], min_length=1, ) @@ -3650,613 +3957,99 @@ class StabilityImageGenrationSD3Response400(BaseModel): examples=['bad_request'], min_length=1, ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) -class StabilityImageGenrationSD3Response413(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) +class Status9(str, Enum): + in_progress = 'in-progress' -class StabilityImageGenrationSD3Response422(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationSD3Response429(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationSD3Response500(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class OutputFormat4(str, Enum): - jpeg = 'jpeg' - png = 'png' - webp = 'webp' - - -class StabilityImageGenrationUpscaleConservativeRequest(BaseModel): - image: StrictBytes = Field( - ..., - description='The image you wish to upscale.\n\nSupported Formats:\n- jpeg\n- png\n- webp\n\nValidation Rules:\n- Every side must be at least 64 pixels\n- Total pixel count must be between 4,096 and 9,437,184 pixels\n- The aspect ratio must be between 1:2.5 and 2.5:1', - examples=['./some/image.png'], - ) - prompt: str = Field( - ..., - description="What you wish to see in the output image. A strong, descriptive prompt that clearly defines\nelements, colors, and subjects will lead to better results.\n\nTo control the weight of a given word use the format `(word:weight)`,\nwhere `word` is the word you'd like to control the weight of and `weight`\nis a value between 0 and 1. For example: `The sky was a crisp (blue:0.3) and (green:0.8)`\nwould convey a sky that was blue and green, but more green than blue.", - max_length=10000, - min_length=1, - ) - negative_prompt: Optional[str] = Field( - None, - description='A blurb of text describing what you **do not** wish to see in the output image.\nThis is an advanced feature.', - max_length=10000, - ) - seed: Optional[float] = Field( - 0, - description="A specific value that is used to guide the 'randomness' of the generation. (Omit this parameter or pass `0` to use a random seed.)", - ge=0.0, - le=4294967294.0, - ) - output_format: Optional[OutputFormat4] = Field( - 'png', description='Dictates the `content-type` of the generated image.' - ) - creativity: Optional[StabilityCreativity] = Field( - default_factory=lambda: StabilityCreativity.model_validate(0.35) - ) - - -class StabilityImageGenrationUpscaleConservativeResponse200(BaseModel): - image: str = Field( - ..., - description='The generated image, encoded to base64.', - examples=['AAAAIGZ0eXBpc29tAAACAGlzb21pc28yYXZjMW1...'], - ) - seed: Optional[float] = Field( - 0, - description='The seed used as random noise for this generation.', - examples=[343940597], - ge=0.0, - le=4294967294.0, - ) - finish_reason: FinishReason = Field( - ..., - description='The reason the generation finished.\n\n- `SUCCESS` = successful generation.\n- `CONTENT_FILTERED` = successful generation, however the output violated our content moderation\npolicy and has been blurred as a result.', - examples=['SUCCESS'], - ) - - -class StabilityImageGenrationUpscaleConservativeResponse400(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleConservativeResponse413(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleConservativeResponse422(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleConservativeResponse429(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleConservativeResponse500(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleCreativeRequest(BaseModel): - image: StrictBytes = Field( - ..., - description='The image you wish to upscale.\n\nSupported Formats:\n- jpeg\n- png\n- webp\n\nValidation Rules:\n- Every side must be at least 64 pixels\n- Total pixel count must be between 4,096 and 1,048,576 pixels', - examples=['./some/image.png'], - ) - prompt: str = Field( - ..., - description="What you wish to see in the output image. A strong, descriptive prompt that clearly defines\nelements, colors, and subjects will lead to better results.\n\nTo control the weight of a given word use the format `(word:weight)`,\nwhere `word` is the word you'd like to control the weight of and `weight`\nis a value between 0 and 1. For example: `The sky was a crisp (blue:0.3) and (green:0.8)`\nwould convey a sky that was blue and green, but more green than blue.", - max_length=10000, - min_length=1, - ) - negative_prompt: Optional[str] = Field( - None, - description='A blurb of text describing what you **do not** wish to see in the output image.\nThis is an advanced feature.', - max_length=10000, - ) - output_format: Optional[OutputFormat4] = Field( - 'png', description='Dictates the `content-type` of the generated image.' - ) - seed: Optional[float] = Field( - 0, - description="A specific value that is used to guide the 'randomness' of the generation. (Omit this parameter or pass `0` to use a random seed.)", - ge=0.0, - le=4294967294.0, - ) - creativity: Optional[float] = Field( - 0.3, - description='Indicates how creative the model should be when upscaling an image.\nHigher values will result in more details being added to the image during upscaling.', - ge=0.1, - le=0.5, - ) - style_preset: Optional[StylePreset] = Field( - None, description='Guides the image model towards a particular style.' - ) - - -class StabilityImageGenrationUpscaleCreativeResponse200(BaseModel): - id: StabilityGenerationID - - -class StabilityImageGenrationUpscaleCreativeResponse400(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleCreativeResponse413(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleCreativeResponse422(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleCreativeResponse429(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleCreativeResponse500(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleFastRequest(BaseModel): - image: StrictBytes = Field( - ..., - description='The image you wish to upscale.\n\nSupported Formats:\n- jpeg\n- png\n- webp\n\nValidation Rules:\n- Width must be between 32 and 1,536 pixels\n- Height must be between 32 and 1,536 pixels\n- Total pixel count must be between 1,024 and 1,048,576 pixels', - examples=['./some/image.png'], - ) - output_format: Optional[OutputFormat4] = Field( - 'png', description='Dictates the `content-type` of the generated image.' - ) - - -class StabilityImageGenrationUpscaleFastResponse200(BaseModel): - image: str = Field( - ..., - description='The generated image, encoded to base64.', - examples=['AAAAIGZ0eXBpc29tAAACAGlzb21pc28yYXZjMW1...'], - ) - seed: Optional[float] = Field( - 0, - description='The seed used as random noise for this generation.', - examples=[343940597], - ge=0.0, - le=4294967294.0, - ) - finish_reason: FinishReason = Field( - ..., - description='The reason the generation finished.\n\n- `SUCCESS` = successful generation.\n- `CONTENT_FILTERED` = successful generation, however the output violated our content moderation\npolicy and has been blurred as a result.', - examples=['SUCCESS'], - ) - - -class StabilityImageGenrationUpscaleFastResponse400(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleFastResponse413(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleFastResponse422(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleFastResponse429(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class StabilityImageGenrationUpscaleFastResponse500(BaseModel): - id: str = Field( - ..., - description='A unique identifier associated with this error. Please include this in any [support tickets](https://kb.stability.ai/knowledge-base/kb-tickets/new)\nyou file, as it will greatly assist us in diagnosing the root cause of the problem.', - examples=['a1b2c3d4e5f6a1b2c3d4e5f6a1b2c3d4'], - min_length=1, - ) - name: str = Field( - ..., - description='Short-hand name for an error, useful for discriminating between errors with the same status code.', - examples=['bad_request'], - min_length=1, - ) - errors: List[str] = Field( - ..., - description='One or more error messages indicating what went wrong.', - examples=[['some-field: is required']], - min_length=1, - ) - - -class ActionJobResult(BaseModel): - id: Optional[UUID] = Field(None, description='Unique identifier for the job result') - workflow_name: Optional[str] = Field(None, description='Name of the workflow') - operating_system: Optional[str] = Field(None, description='Operating system used') - python_version: Optional[str] = Field(None, description='PyTorch version used') - pytorch_version: Optional[str] = Field(None, description='PyTorch version used') - action_run_id: Optional[str] = Field( - None, description='Identifier of the run this result belongs to' - ) - action_job_id: Optional[str] = Field( - None, description='Identifier of the job this result belongs to' - ) - cuda_version: Optional[str] = Field(None, description='CUDA version used') - branch_name: Optional[str] = Field( - None, description='Name of the relevant git branch' - ) - commit_hash: Optional[str] = Field(None, description='The hash of the commit') - commit_id: Optional[str] = Field(None, description='The ID of the commit') - commit_time: Optional[int] = Field( - None, description='The Unix timestamp when the commit was made' - ) - commit_message: Optional[str] = Field(None, description='The message of the commit') - comfy_run_flags: Optional[str] = Field( - None, description='The comfy run flags. E.g. `--low-vram`' - ) - git_repo: Optional[str] = Field(None, description='The repository name') - pr_number: Optional[str] = Field(None, description='The pull request number') - start_time: Optional[int] = Field( - None, description='The start time of the job as a Unix timestamp.' - ) - end_time: Optional[int] = Field( - None, description='The end time of the job as a Unix timestamp.' - ) - avg_vram: Optional[int] = Field( - None, description='The average VRAM used by the job' - ) - peak_vram: Optional[int] = Field(None, description='The peak VRAM used by the job') - job_trigger_user: Optional[str] = Field( - None, description='The user who triggered the job.' - ) - author: Optional[str] = Field(None, description='The author of the commit') - machine_stats: Optional[MachineStats] = None - status: Optional[WorkflowRunStatus] = None - storage_file: Optional[StorageFile] = None - - -class Publisher(BaseModel): - name: Optional[str] = None +class StabilityGetResultResponse202(BaseModel): id: Optional[str] = Field( + None, description='The ID of the generation result.', examples=[1234567890] + ) + status: Optional[Status9] = None + + +class Type20(str, Enum): + json_schema = 'json_schema' + + +class TextResponseFormatJsonSchema(BaseModel): + description: Optional[str] = Field( None, - description="The unique identifier for the publisher. It's akin to a username. Should be lowercase.", + description='A description of what the response format is for, used by the model to\ndetermine how to respond in the format.\n', ) - description: Optional[str] = None - website: Optional[str] = None - support: Optional[str] = None - source_code_repo: Optional[str] = None - logo: Optional[str] = Field(None, description="URL to the publisher's logo.") - createdAt: Optional[datetime] = Field( - None, description='The date and time the publisher was created.' + name: str = Field( + ..., + description='The name of the response format. Must be a-z, A-Z, 0-9, or contain\nunderscores and dashes, with a maximum length of 64.\n', ) - members: Optional[List[PublisherMember]] = Field( - None, description='A list of members in the publisher.' + schema_: ResponseFormatJsonSchemaSchema = Field(..., alias='schema') + strict: Optional[bool] = Field( + False, + description='Whether to enable strict schema adherence when generating the output.\nIf set to true, the model will always follow the exact schema defined\nin the `schema` field. Only a subset of JSON Schema is supported when\n`strict` is `true`. To learn more, read the [Structured Outputs\nguide](/docs/guides/structured-outputs).\n', ) - status: Optional[PublisherStatus] = Field( - None, description='The status of the publisher.' + type: Type20 = Field( + ..., + description='The type of response format being defined. Always `json_schema`.', ) -class NodeVersion(BaseModel): - id: Optional[str] = None - version: Optional[str] = Field( +class Type21(str, Enum): + function = 'function' + + +class ToolChoiceFunction(BaseModel): + name: str = Field(..., description='The name of the function to call.') + type: Type21 = Field( + ..., description='For function calling, the type is always `function`.' + ) + + +class ToolChoiceOptions(str, Enum): + none = 'none' + auto = 'auto' + required = 'required' + + +class Type22(str, Enum): + file_search = 'file_search' + web_search_preview = 'web_search_preview' + computer_use_preview = 'computer_use_preview' + web_search_preview_2025_03_11 = 'web_search_preview_2025_03_11' + + +class ToolChoiceTypes(BaseModel): + type: Type22 = Field( + ..., + description='The type of hosted tool the model should to use. Learn more about\n[built-in tools](/docs/guides/tools).\n\nAllowed values are:\n- `file_search`\n- `web_search_preview`\n- `computer_use_preview`\n', + ) + + +class Veo2GenVidPollRequest(BaseModel): + operationName: str = Field( + ..., + description='Full operation name (from predict response)', + examples=[ + 'projects/PROJECT_ID/locations/us-central1/publishers/google/models/MODEL_ID/operations/OPERATION_ID' + ], + ) + + +class Error(BaseModel): + code: Optional[int] = Field(None, description='Error code') + message: Optional[str] = Field(None, description='Error message') + + +class Video(BaseModel): + bytesBase64Encoded: Optional[str] = Field( + None, description='Base64-encoded video content' + ) + gcsUri: Optional[str] = Field(None, description='Cloud Storage URI of the video') + mimeType: Optional[str] = Field(None, description='Video MIME type') + + +class Response(BaseModel): + field_type: Optional[str] = Field( None, description='The number of images to generate (1-10). Only 1 supported for dall-e-3.', examples=[1], @@ -4292,80 +4085,150 @@ class NodeVersion(BaseModel): description='A unique identifier for end-user monitoring', examples=['user-1234'], ) - createdAt: Optional[datetime] = Field( - None, description='The date and time the version was created.' + raiMediaFilteredCount: Optional[int] = Field( + None, description='Count of media filtered by responsible AI policies' ) - changelog: Optional[str] = Field( - None, description='Summary of changes made in this version' + raiMediaFilteredReasons: Optional[List[str]] = Field( + None, description='Reasons why media was filtered by responsible AI policies' ) - dependencies: Optional[List[str]] = Field( - None, description='A list of pip dependencies required by the node.' + videos: Optional[List[Video]] = None + + +class Veo2GenVidPollResponse(BaseModel): + done: Optional[bool] = None + error: Optional[Error] = Field( + None, description='Error details if operation failed' ) - downloadUrl: Optional[str] = Field( - None, description='[Output Only] URL to download this version of the node' - ) - deprecated: Optional[bool] = Field( - None, description='Indicates if this version is deprecated.' - ) - status: Optional[NodeVersionStatus] = Field( - None, description='The status of the node version.' - ) - status_reason: Optional[str] = Field( - None, description='The reason for the status change.' - ) - node_id: Optional[str] = Field( - None, description='The unique identifier of the node.' - ) - comfy_node_extract_status: Optional[str] = Field( - None, description='The status of comfy node extraction process.' + name: Optional[str] = None + response: Optional[Response] = Field( + None, description='The actual prediction response if done is true' ) -class IdeogramV3Request(BaseModel): - prompt: str = Field(..., description='The text prompt for image generation') - seed: Optional[int] = Field( - None, description='Seed value for reproducible generation' +class Image(BaseModel): + bytesBase64Encoded: str + gcsUri: Optional[str] = None + mimeType: Optional[str] = None + + +class Image1(BaseModel): + bytesBase64Encoded: Optional[str] = None + gcsUri: str + mimeType: Optional[str] = None + + +class Instance(BaseModel): + image: Optional[Union[Image, Image1]] = Field( + None, description='Optional image to guide video generation' ) - resolution: Optional[str] = Field( - None, description='Image resolution in format WxH', examples=['1280x800'] + prompt: str = Field(..., description='Text description of the video') + + +class PersonGeneration1(str, Enum): + ALLOW = 'ALLOW' + BLOCK = 'BLOCK' + + +class Parameters(BaseModel): + aspectRatio: Optional[str] = Field(None, examples=['16:9']) + durationSeconds: Optional[int] = None + enhancePrompt: Optional[bool] = None + negativePrompt: Optional[str] = None + personGeneration: Optional[PersonGeneration1] = None + sampleCount: Optional[int] = None + seed: Optional[int] = None + storageUri: Optional[str] = Field( + None, description='Optional Cloud Storage URI to upload the video' ) - aspect_ratio: Optional[str] = Field( - None, description='Aspect ratio in format WxH', examples=['1x3'] + + +class Veo2GenVidRequest(BaseModel): + instances: Optional[List[Instance]] = None + parameters: Optional[Parameters] = None + + +class Veo2GenVidResponse(BaseModel): + name: str = Field( + ..., + description='Operation resource name', + examples=[ + 'projects/PROJECT_ID/locations/us-central1/publishers/google/models/MODEL_ID/operations/a1b07c8e-7b5a-4aba-bb34-3e1ccb8afcc8' + ], ) - rendering_speed: RenderingSpeed - magic_prompt: Optional[MagicPrompt] = Field( - None, description='Whether to enable magic prompt enhancement' + + +class SearchContextSize(str, Enum): + low = 'low' + medium = 'medium' + high = 'high' + + +class Type23(str, Enum): + web_search_preview = 'web_search_preview' + web_search_preview_2025_03_11 = 'web_search_preview_2025_03_11' + + +class WebSearchPreviewTool(BaseModel): + search_context_size: Optional[SearchContextSize] = Field( + None, + description='High level guidance for the amount of context window space to use for the search. One of `low`, `medium`, or `high`. `medium` is the default.', ) - negative_prompt: Optional[str] = Field( - None, description='Text prompt specifying what to avoid in the generation' + type: Literal['WebSearchPreviewTool'] = Field( + ..., + description='The type of the web search tool. One of `web_search_preview` or `web_search_preview_2025_03_11`.', ) - num_images: Optional[int] = Field( - None, description='Number of images to generate', ge=1 + + +class Status10(str, Enum): + in_progress = 'in_progress' + searching = 'searching' + completed = 'completed' + failed = 'failed' + + +class Type24(str, Enum): + web_search_call = 'web_search_call' + + +class WebSearchToolCall(BaseModel): + id: str = Field(..., description='The unique ID of the web search tool call.\n') + status: Status10 = Field( + ..., description='The status of the web search tool call.\n' ) - color_palette: Optional[ColorPalette] = None - style_codes: Optional[List[StyleCode]] = Field( - None, description='Array of style codes in hexadecimal format' - ) - style_type: Optional[StyleType] = Field( - None, description='The type of style to apply' - ) - style_reference_images: Optional[List[str]] = Field( - None, description='Array of reference image URLs or identifiers' + type: Type24 = Field( + ..., + description='The type of the web search tool call. Always `web_search_call`.\n', ) +CreateModelResponseProperties = ModelResponseProperties + + +class GeminiContent(BaseModel): + parts: List[GeminiPart] + role: str + + +class GeminiGenerateContentRequest(BaseModel): + contents: List[GeminiContent] + generationConfig: Optional[GeminiGenerationConfig] = None + safetySettings: Optional[List[GeminiSafetySetting]] = None + systemInstruction: Optional[GeminiContent] = None + tools: Optional[List[GeminiTool]] = None + + +class GeminiPromptFeedback(BaseModel): + blockReason: Optional[str] = None + blockReasonMessage: Optional[str] = None + safetyRatings: Optional[List[GeminiSafetyRating]] = None + + class IdeogramV3EditRequest(BaseModel): + color_palette: Optional[IdeogramColorPalette] = None image: Optional[StrictBytes] = Field( None, description='The image being edited (max size 10MB); only JPEG, WebP and PNG formats are supported at this time.', ) - mask: Optional[StrictBytes] = Field( - None, - description='A black and white image of the same size as the image being edited (max size 10MB). Black regions in the mask should match up with the regions of the image that you would like to edit; only JPEG, WebP and PNG formats are supported at this time.', - ) - prompt: str = Field( - ..., description='The prompt used to describe the edited result.' - ) magic_prompt: Optional[str] = Field( None, description='Determine if MagicPrompt should be used in generating the request or not.', @@ -4384,13 +4247,12 @@ class Response(BaseModel): num_images: Optional[int] = Field( None, description='The number of images to generate.' ) - seed: Optional[int] = Field( - None, description='Random seed. Set for reproducible generation.' + prompt: str = Field( + ..., description='The prompt used to describe the edited result.' ) rendering_speed: RenderingSpeed - color_palette: Optional[IdeogramColorPalette] = Field( - None, - description='A color palette for generation, must EITHER be specified via one of the presets (name) or explicitly via hexadecimal representations of the color with optional weights (members). Not supported by V_1, V_1_TURBO, V_2A and V_2A_TURBO models.', + seed: Optional[int] = Field( + None, description='Random seed. Set for reproducible generation.' ) style_codes: Optional[List[StyleCode]] = Field( None, @@ -4405,34 +4267,102 @@ class Response(BaseModel): duration: LumaVideoModelOutputDuration -class KlingCameraControl(BaseModel): - type: Optional[KlingCameraControlType] = None - config: Optional[KlingCameraConfig] = None - - -class KlingText2VideoRequest(BaseModel): - model_name: Optional[KlingVideoGenModelName] = 'kling-v2-master' - prompt: Optional[str] = Field( - None, description='Positive text prompt', max_length=2500 +class IdeogramV3Request(BaseModel): + aspect_ratio: Optional[str] = Field( + None, description='Aspect ratio in format WxH', examples=['1x3'] + ) + color_palette: Optional[ColorPalette] = None + magic_prompt: Optional[MagicPrompt2] = Field( + None, description='Whether to enable magic prompt enhancement' ) negative_prompt: Optional[str] = Field( - None, description='Negative text prompt', max_length=2500 + None, description='Text prompt specifying what to avoid in the generation' ) - cfg_scale: Optional[KlingVideoGenCfgScale] = Field( - default_factory=lambda: KlingVideoGenCfgScale.model_validate(0.5) + num_images: Optional[int] = Field( + None, description='Number of images to generate', ge=1 ) + prompt: str = Field(..., description='The text prompt for image generation') + rendering_speed: RenderingSpeed + resolution: Optional[str] = Field( + None, description='Image resolution in format WxH', examples=['1280x800'] + ) + seed: Optional[int] = Field( + None, description='Seed value for reproducible generation' + ) + style_codes: Optional[List[StyleCode]] = Field( + None, description='Array of style codes in hexadecimal format' + ) + style_reference_images: Optional[List[str]] = Field( + None, description='Array of reference image URLs or identifiers' + ) + style_type: Optional[StyleType1] = Field( + None, description='The type of style to apply' + ) + + +class ImagenGenerateImageResponse(BaseModel): + predictions: Optional[List[ImagenImagePrediction]] = None + + +class ImagenImageGenerationParameters(BaseModel): + addWatermark: Optional[bool] = None + aspectRatio: Optional[AspectRatio] = None + enhancePrompt: Optional[bool] = None + includeRaiReason: Optional[bool] = None + includeSafetyAttributes: Optional[bool] = None + outputOptions: Optional[ImagenOutputOptions] = None + personGeneration: Optional[PersonGeneration] = None + safetySetting: Optional[SafetySetting] = None + sampleCount: Optional[int] = Field(None, ge=1, le=4) + seed: Optional[int] = None + storageUri: Optional[AnyUrl] = None + + +class InputContent( + RootModel[Union[InputTextContent, InputImageContent, InputFileContent]] +): + root: Union[InputTextContent, InputImageContent, InputFileContent] + + +class InputMessageContentList(RootModel[List[InputContent]]): + root: List[InputContent] = Field( + ..., + description='A list of one or many input items to the model, containing different content \ntypes.\n', + title='Input item content list', + ) + + +class KlingCameraControl(BaseModel): + config: Optional[KlingCameraConfig] = None + type: Optional[KlingCameraControlType] = None + + +class KlingDualCharacterEffectInput(BaseModel): + duration: KlingVideoGenDuration + images: KlingDualCharacterImages mode: Optional[KlingVideoGenMode] = 'std' - camera_control: Optional[KlingCameraControl] = None - aspect_ratio: Optional[KlingVideoGenAspectRatio] = '16:9' - duration: Optional[KlingVideoGenDuration] = '5' - callback_url: Optional[AnyUrl] = Field( - None, description='The callback notification address' - ) - external_task_id: Optional[str] = Field(None, description='Customized Task ID') + model_name: Optional[KlingCharacterEffectModelName] = 'kling-v1' class KlingImage2VideoRequest(BaseModel): - model_name: Optional[KlingVideoGenModelName] = 'kling-v2-master' + aspect_ratio: Optional[KlingVideoGenAspectRatio] = '16:9' + callback_url: Optional[AnyUrl] = Field( + None, + description='The callback notification address. Server will notify when the task status changes.', + ) + camera_control: Optional[KlingCameraControl] = None + cfg_scale: Optional[KlingVideoGenCfgScale] = Field( + default_factory=lambda: KlingVideoGenCfgScale.model_validate(0.5) + ) + duration: Optional[KlingVideoGenDuration] = '5' + dynamic_masks: Optional[List[DynamicMask]] = Field( + None, + description='Dynamic Brush Configuration List (up to 6 groups). For 5-second videos, trajectory length must not exceed 77 coordinates.', + ) + external_task_id: Optional[str] = Field( + None, + description='Customized Task ID. Must be unique within a single user account.', + ) image: Optional[str] = Field( None, description='Reference Image - URL or Base64 encoded string, cannot exceed 10MB, resolution not less than 300*300px, aspect ratio between 1:2.5 ~ 2.5:1. Base64 should not include data:image prefix.', @@ -4441,35 +4371,168 @@ class KlingImage2VideoRequest(BaseModel): None, description='Reference Image - End frame control. URL or Base64 encoded string, cannot exceed 10MB, resolution not less than 300*300px. Base64 should not include data:image prefix.', ) - prompt: Optional[str] = Field( - None, description='Positive text prompt', max_length=2500 - ) + mode: Optional[KlingVideoGenMode] = 'std' + model_name: Optional[KlingVideoGenModelName] = 'kling-v2-master' negative_prompt: Optional[str] = Field( None, description='Negative text prompt', max_length=2500 ) - cfg_scale: Optional[KlingVideoGenCfgScale] = Field( - default_factory=lambda: KlingVideoGenCfgScale.model_validate(0.5) + prompt: Optional[str] = Field( + None, description='Positive text prompt', max_length=2500 ) - mode: Optional[KlingVideoGenMode] = 'std' static_mask: Optional[str] = Field( None, description='Static Brush Application Area (Mask image created by users using the motion brush). The aspect ratio must match the input image.', ) - dynamic_masks: Optional[List[DynamicMask]] = Field( + + +class TaskResult(BaseModel): + videos: Optional[List[KlingVideoResult]] = None + + +class Data(BaseModel): + created_at: Optional[int] = Field(None, description='Task creation time') + task_id: Optional[str] = Field(None, description='Task ID') + task_info: Optional[TaskInfo] = None + task_result: Optional[TaskResult] = None + task_status: Optional[KlingTaskStatus] = None + updated_at: Optional[int] = Field(None, description='Task update time') + + +class KlingImage2VideoResponse(BaseModel): + code: Optional[int] = Field(None, description='Error code') + data: Optional[Data] = None + message: Optional[str] = Field(None, description='Error message') + request_id: Optional[str] = Field(None, description='Request ID') + + +class TaskResult1(BaseModel): + images: Optional[List[KlingImageResult]] = None + + +class Data1(BaseModel): + created_at: Optional[int] = Field(None, description='Task creation time') + task_id: Optional[str] = Field(None, description='Task ID') + task_result: Optional[TaskResult1] = None + task_status: Optional[KlingTaskStatus] = None + task_status_msg: Optional[str] = Field(None, description='Task status information') + updated_at: Optional[int] = Field(None, description='Task update time') + + +class KlingImageGenerationsResponse(BaseModel): + code: Optional[int] = Field(None, description='Error code') + data: Optional[Data1] = None + message: Optional[str] = Field(None, description='Error message') + request_id: Optional[str] = Field(None, description='Request ID') + + +class KlingLipSyncInputObject(BaseModel): + audio_file: Optional[str] = Field( None, - description='Dynamic Brush Configuration List (up to 6 groups). For 5-second videos, trajectory length must not exceed 77 coordinates.', + description='Local Path of Audio File. Supported formats: .mp3/.wav/.m4a/.aac, maximum file size of 5MB. Base64 code.', ) - camera_control: Optional[KlingCameraControl] = None - aspect_ratio: Optional[KlingVideoGenAspectRatio] = '16:9' - duration: Optional[KlingVideoGenDuration] = '5' + audio_type: Optional[KlingAudioUploadType] = None + audio_url: Optional[str] = Field( + None, + description='Audio File Download URL. Supported formats: .mp3/.wav/.m4a/.aac, maximum file size of 5MB.', + ) + mode: KlingLipSyncMode + text: Optional[str] = Field( + None, + description='Text Content for Lip-Sync Video Generation. Required when mode is text2video. Maximum length is 120 characters.', + ) + video_id: Optional[str] = Field( + None, + description='The ID of the video generated by Kling AI. Only supports 5-second and 10-second videos generated within the last 30 days.', + ) + video_url: Optional[str] = Field( + None, + description='Get link for uploaded video. Video files support .mp4/.mov, file size does not exceed 100MB, video length between 2-10s.', + ) + voice_id: Optional[str] = Field( + None, + description='Voice ID. Required when mode is text2video. The system offers a variety of voice options to choose from.', + ) + voice_language: Optional[KlingLipSyncVoiceLanguage] = 'en' + voice_speed: Optional[float] = Field( + 1, + description='Speech Rate. Valid range: 0.8~2.0, accurate to one decimal place.', + ge=0.8, + le=2.0, + ) + + +class KlingLipSyncRequest(BaseModel): callback_url: Optional[AnyUrl] = Field( None, description='The callback notification address. Server will notify when the task status changes.', ) - external_task_id: Optional[str] = Field( - None, - description='Customized Task ID. Must be unique within a single user account.', + input: KlingLipSyncInputObject + + +class TaskResult2(BaseModel): + videos: Optional[List[KlingVideoResult]] = None + + +class Data2(BaseModel): + created_at: Optional[int] = Field(None, description='Task creation time') + task_id: Optional[str] = Field(None, description='Task ID') + task_info: Optional[TaskInfo] = None + task_result: Optional[TaskResult2] = None + task_status: Optional[KlingTaskStatus] = None + updated_at: Optional[int] = Field(None, description='Task update time') + + +class KlingLipSyncResponse(BaseModel): + code: Optional[int] = Field(None, description='Error code') + data: Optional[Data2] = None + message: Optional[str] = Field(None, description='Error message') + request_id: Optional[str] = Field(None, description='Request ID') + + +class KlingSingleImageEffectInput(BaseModel): + duration: KlingSingleImageEffectDuration + image: str = Field( + ..., + description='Reference Image. URL or Base64 encoded string (without data:image prefix). File size cannot exceed 10MB, resolution not less than 300*300px, aspect ratio between 1:2.5 ~ 2.5:1.', ) + model_name: KlingSingleImageEffectModelName + + +class KlingText2VideoRequest(BaseModel): + aspect_ratio: Optional[KlingVideoGenAspectRatio] = '16:9' + callback_url: Optional[AnyUrl] = Field( + None, description='The callback notification address' + ) + camera_control: Optional[KlingCameraControl] = None + cfg_scale: Optional[KlingVideoGenCfgScale] = Field( + default_factory=lambda: KlingVideoGenCfgScale.model_validate(0.5) + ) + duration: Optional[KlingVideoGenDuration] = '5' + external_task_id: Optional[str] = Field(None, description='Customized Task ID') + mode: Optional[KlingVideoGenMode] = 'std' + model_name: Optional[KlingTextToVideoModelName] = 'kling-v1' + negative_prompt: Optional[str] = Field( + None, description='Negative text prompt', max_length=2500 + ) + prompt: Optional[str] = Field( + None, description='Positive text prompt', max_length=2500 + ) + + +class Data4(BaseModel): + created_at: Optional[int] = Field(None, description='Task creation time') + task_id: Optional[str] = Field(None, description='Task ID') + task_info: Optional[TaskInfo] = None + task_result: Optional[TaskResult2] = None + task_status: Optional[KlingTaskStatus] = None + updated_at: Optional[int] = Field(None, description='Task update time') + + +class KlingText2VideoResponse(BaseModel): + code: Optional[int] = Field(None, description='Error code') + data: Optional[Data4] = None + message: Optional[str] = Field(None, description='Error message') + request_id: Optional[str] = Field(None, description='Request ID') class KlingVideoEffectsInput( @@ -4478,219 +4541,66 @@ class KlingVideoEffectsInput( root: Union[KlingSingleImageEffectInput, KlingDualCharacterEffectInput] -class StripeBillingDetails(BaseModel): - address: Optional[StripeAddress] = None - email: Optional[str] = None - name: Optional[str] = None - phone: Optional[str] = None - tax_id: Optional[Any] = None - - -class StripePaymentMethodDetails(BaseModel): - card: Optional[StripeCardDetails] = None - type: Optional[str] = None - - -class BFLFluxProFillInputs(BaseModel): - image: str = Field( - ..., - description='A Base64-encoded string representing the image you wish to modify. Can contain alpha mask if desired.', - title='Image', - ) - mask: Optional[str] = Field( +class KlingVideoEffectsRequest(BaseModel): + callback_url: Optional[AnyUrl] = Field( None, - description='A Base64-encoded string representing a mask for the areas you want to modify in the image. The mask should be the same dimensions as the image and in black and white. Black areas (0%) indicate no modification, while white areas (100%) specify areas for inpainting. Optional if you provide an alpha mask in the original image. Validation: The endpoint verifies that the dimensions of the mask match the original image.', - title='Mask', + description='The callback notification address for the result of this task.', + ) + effect_scene: Union[KlingDualCharacterEffectsScene, KlingSingleImageEffectsScene] + external_task_id: Optional[str] = Field( + None, + description='Customized Task ID. Must be unique within a single user account.', + ) + input: KlingVideoEffectsInput + + +class Data5(BaseModel): + created_at: Optional[int] = Field(None, description='Task creation time') + task_id: Optional[str] = Field(None, description='Task ID') + task_info: Optional[TaskInfo] = None + task_result: Optional[TaskResult2] = None + task_status: Optional[KlingTaskStatus] = None + updated_at: Optional[int] = Field(None, description='Task update time') + + +class KlingVideoEffectsResponse(BaseModel): + code: Optional[int] = Field(None, description='Error code') + data: Optional[Data5] = None + message: Optional[str] = Field(None, description='Error message') + request_id: Optional[str] = Field(None, description='Request ID') + + +class KlingVideoExtendRequest(BaseModel): + callback_url: Optional[AnyUrl] = Field( + None, + description='The callback notification address. Server will notify when the task status changes.', + ) + cfg_scale: Optional[KlingVideoGenCfgScale] = Field( + default_factory=lambda: KlingVideoGenCfgScale.model_validate(0.5) + ) + negative_prompt: Optional[str] = Field( + None, + description='Negative text prompt for elements to avoid in the extended video', + max_length=2500, ) prompt: Optional[str] = Field( - '', - description='The description of the changes you want to make. This text guides the inpainting process, allowing you to specify features, styles, or modifications for the masked area.', - examples=['ein fantastisches bild'], - title='Prompt', - ) - steps: Optional[Steps] = Field( - default_factory=lambda: Steps.model_validate(50), - description='Number of steps for the image generation process', - examples=[50], - title='Steps', - ) - prompt_upsampling: Optional[bool] = Field( - False, - description='Whether to perform upsampling on the prompt. If active, automatically modifies the prompt for more creative generation', - title='Prompt Upsampling', - ) - seed: Optional[int] = Field( - None, description='Optional seed for reproducibility', title='Seed' - ) - guidance: Optional[Guidance] = Field( - default_factory=lambda: Guidance.model_validate(60), - description='Guidance strength for the image generation process', - title='Guidance', - ) - output_format: Optional[BFLOutputFormat] = Field( - 'jpeg', - description="Output format for the generated image. Can be 'jpeg' or 'png'.", - ) - safety_tolerance: Optional[int] = Field( - 2, - description='Tolerance level for input and output moderation. Between 0 and 6, 0 being most strict, 6 being least strict.', - examples=[2], - ge=0, - le=6, - title='Safety Tolerance', - ) - webhook_url: Optional[WebhookUrl] = Field( - None, description='URL to receive webhook notifications', title='Webhook Url' - ) - webhook_secret: Optional[str] = Field( None, - description='Optional secret for webhook signature verification', - title='Webhook Secret', + description='Positive text prompt for guiding the video extension', + max_length=2500, + ) + video_id: Optional[str] = Field( + None, + description='The ID of the video to be extended. Supports videos generated by text-to-video, image-to-video, and previous video extension operations. Cannot exceed 3 minutes total duration after extension.', ) -class BFLHTTPValidationError(BaseModel): - detail: Optional[List[BFLValidationError]] = Field(None, title='Detail') - - -class BFLFluxProExpandInputs(BaseModel): - image: str = Field( - ..., - description='A Base64-encoded string representing the image you wish to expand.', - title='Image', - ) - top: Optional[Top] = Field( - 0, description='Number of pixels to expand at the top of the image', title='Top' - ) - bottom: Optional[Bottom] = Field( - 0, - description='Number of pixels to expand at the bottom of the image', - title='Bottom', - ) - left: Optional[Left] = Field( - 0, - description='Number of pixels to expand on the left side of the image', - title='Left', - ) - right: Optional[Right] = Field( - 0, - description='Number of pixels to expand on the right side of the image', - title='Right', - ) - prompt: Optional[str] = Field( - '', - description='The description of the changes you want to make. This text guides the expansion process, allowing you to specify features, styles, or modifications for the expanded areas.', - examples=['ein fantastisches bild'], - title='Prompt', - ) - steps: Optional[Steps] = Field( - default_factory=lambda: Steps.model_validate(50), - description='Number of steps for the image generation process', - examples=[50], - title='Steps', - ) - prompt_upsampling: Optional[bool] = Field( - False, - description='Whether to perform upsampling on the prompt. If active, automatically modifies the prompt for more creative generation', - title='Prompt Upsampling', - ) - seed: Optional[int] = Field( - None, description='Optional seed for reproducibility', title='Seed' - ) - guidance: Optional[Guidance] = Field( - default_factory=lambda: Guidance.model_validate(60), - description='Guidance strength for the image generation process', - title='Guidance', - ) - output_format: Optional[BFLOutputFormat] = Field( - 'jpeg', - description="Output format for the generated image. Can be 'jpeg' or 'png'.", - ) - safety_tolerance: Optional[int] = Field( - 2, - description='Tolerance level for input and output moderation. Between 0 and 6, 0 being most strict, 6 being least strict.', - examples=[2], - ge=0, - le=6, - title='Safety Tolerance', - ) - webhook_url: Optional[WebhookUrl] = Field( - None, description='URL to receive webhook notifications', title='Webhook Url' - ) - webhook_secret: Optional[str] = Field( - None, - description='Optional secret for webhook signature verification', - title='Webhook Secret', - ) - - -class BFLCannyInputs(BaseModel): - prompt: str = Field( - ..., - description='Text prompt for image generation', - examples=['ein fantastisches bild'], - title='Prompt', - ) - control_image: Optional[str] = Field( - None, - description='Base64 encoded image to use as control input if no preprocessed image is provided', - title='Control Image', - ) - preprocessed_image: Optional[str] = Field( - None, - description='Optional pre-processed image that will bypass the control preprocessing step', - title='Preprocessed Image', - ) - canny_low_threshold: Optional[CannyLowThreshold] = Field( - default_factory=lambda: CannyLowThreshold.model_validate(50), - description='Low threshold for Canny edge detection', - title='Canny Low Threshold', - ) - canny_high_threshold: Optional[CannyHighThreshold] = Field( - default_factory=lambda: CannyHighThreshold.model_validate(200), - description='High threshold for Canny edge detection', - title='Canny High Threshold', - ) - prompt_upsampling: Optional[bool] = Field( - False, - description='Whether to perform upsampling on the prompt', - title='Prompt Upsampling', - ) - seed: Optional[int] = Field( - None, - description='Optional seed for reproducibility', - examples=[42], - title='Seed', - ) - steps: Optional[Steps2] = Field( - default_factory=lambda: Steps2.model_validate(50), - description='Number of steps for the image generation process', - title='Steps', - ) - output_format: Optional[BFLOutputFormat] = Field( - 'jpeg', - description="Output format for the generated image. Can be 'jpeg' or 'png'.", - ) - guidance: Optional[Guidance2] = Field( - default_factory=lambda: Guidance2.model_validate(30), - description='Guidance strength for the image generation process', - title='Guidance', - ) - safety_tolerance: Optional[int] = Field( - 2, - description='Tolerance level for input and output moderation. Between 0 and 6, 0 being most strict, 6 being least strict.', - ge=0, - le=6, - title='Safety Tolerance', - ) - webhook_url: Optional[WebhookUrl] = Field( - None, description='URL to receive webhook notifications', title='Webhook Url' - ) - webhook_secret: Optional[str] = Field( - None, - description='Optional secret for webhook signature verification', - title='Webhook Secret', - ) +class Data6(BaseModel): + created_at: Optional[int] = Field(None, description='Task creation time') + task_id: Optional[str] = Field(None, description='Task ID') + task_info: Optional[TaskInfo] = None + task_result: Optional[TaskResult2] = None + task_status: Optional[KlingTaskStatus] = None + updated_at: Optional[int] = Field(None, description='Task update time') class BFLDepthInputs(BaseModel): @@ -4796,18 +4706,220 @@ class LumaKeyframes(BaseModel): class LumaGenerationRequest(BaseModel): - generation_type: Optional[GenerationType] = 'video' - prompt: str = Field(..., description='The prompt of the generation') aspect_ratio: LumaAspectRatio - loop: Optional[bool] = Field(None, description='Whether to loop the video') - keyframes: Optional[LumaKeyframes] = None callback_url: Optional[AnyUrl] = Field( None, description='The callback URL of the generation, a POST request with Generation object will be sent to the callback URL when the generation is dreaming, completed, or failed', ) - model: LumaVideoModel - resolution: LumaVideoModelOutputResolution duration: LumaVideoModelOutputDuration + generation_type: Optional[GenerationType1] = 'video' + keyframes: Optional[LumaKeyframes] = None + loop: Optional[bool] = Field(None, description='Whether to loop the video') + model: LumaVideoModel + prompt: str = Field(..., description='The prompt of the generation') + resolution: LumaVideoModelOutputResolution + + +class CharacterRef(BaseModel): + identity0: Optional[LumaImageIdentity] = None + + +class LumaImageGenerationRequest(BaseModel): + aspect_ratio: Optional[LumaAspectRatio] = '16:9' + callback_url: Optional[AnyUrl] = Field( + None, description='The callback URL for the generation' + ) + character_ref: Optional[CharacterRef] = None + generation_type: Optional[GenerationType2] = 'image' + image_ref: Optional[List[LumaImageRef]] = None + model: Optional[LumaImageModel] = 'photon-1' + modify_image_ref: Optional[LumaModifyImageRef] = None + prompt: Optional[str] = Field(None, description='The prompt of the generation') + style_ref: Optional[List[LumaImageRef]] = None + + +class LumaUpscaleVideoGenerationRequest(BaseModel): + callback_url: Optional[AnyUrl] = Field( + None, description='The callback URL for the upscale' + ) + generation_type: Optional[GenerationType3] = 'upscale_video' + resolution: Optional[LumaVideoModelOutputResolution] = None + + +class OutputContent(RootModel[Union[OutputTextContent, OutputAudioContent]]): + root: Union[OutputTextContent, OutputAudioContent] + + +class OutputMessage(BaseModel): + content: List[OutputContent] = Field(..., description='The content of the message') + role: Role2 = Field(..., description='The role of the message') + type: Type14 = Field(..., description='The type of output item') + + +class PikaBodyGenerate22I2vGenerate22I2vPost(BaseModel): + duration: Optional[PikaDurationEnum] = 5 + image: Optional[StrictBytes] = Field(None, title='Image') + negativePrompt: Optional[str] = Field(None, title='Negativeprompt') + promptText: Optional[str] = Field(None, title='Prompttext') + resolution: Optional[PikaResolutionEnum] = '1080p' + seed: Optional[int] = Field(None, title='Seed') + + +class PikaBodyGenerate22KeyframeGenerate22PikaframesPost(BaseModel): + duration: Optional[int] = Field(None, ge=5, le=10, title='Duration') + keyFrames: Optional[List[StrictBytes]] = Field( + None, description='Array of keyframe images', title='Keyframes' + ) + negativePrompt: Optional[str] = Field(None, title='Negativeprompt') + promptText: str = Field(..., title='Prompttext') + resolution: Optional[PikaResolutionEnum] = '1080p' + seed: Optional[int] = Field(None, title='Seed') + + +class PikaBodyGenerate22T2vGenerate22T2vPost(BaseModel): + aspectRatio: Optional[float] = Field( + 1.7777777777777777, + description='Aspect ratio (width / height)', + ge=0.4, + le=2.5, + title='Aspectratio', + ) + duration: Optional[PikaDurationEnum] = 5 + negativePrompt: Optional[str] = Field(None, title='Negativeprompt') + promptText: str = Field(..., title='Prompttext') + resolution: Optional[PikaResolutionEnum] = '1080p' + seed: Optional[int] = Field(None, title='Seed') + + +class PikaBodyGeneratePikaffectsGeneratePikaffectsPost(BaseModel): + image: Optional[StrictBytes] = Field(None, title='Image') + negativePrompt: Optional[str] = Field(None, title='Negativeprompt') + pikaffect: Optional[Pikaffect] = None + promptText: Optional[str] = Field(None, title='Prompttext') + seed: Optional[int] = Field(None, title='Seed') + + +class PikaHTTPValidationError(BaseModel): + detail: Optional[List[PikaValidationError]] = Field(None, title='Detail') + + +class Reasoning(BaseModel): + effort: Optional[ReasoningEffort] = 'medium' + generate_summary: Optional[GenerateSummary] = Field( + None, + description="**Deprecated:** use `summary` instead.\n\nA summary of the reasoning performed by the model. This can be\nuseful for debugging and understanding the model's reasoning process.\nOne of `auto`, `concise`, or `detailed`.\n", + ) + summary: Optional[Summary] = Field( + None, + description="A summary of the reasoning performed by the model. This can be\nuseful for debugging and understanding the model's reasoning process.\nOne of `auto`, `concise`, or `detailed`.\n", + ) + + +class ResponseError(BaseModel): + code: ResponseErrorCode + message: str = Field(..., description='A human-readable description of the error.') + + +class RunwayImageToVideoRequest(BaseModel): + duration: RunwayDurationEnum + model: RunwayModelEnum + promptImage: RunwayPromptImageObject + promptText: Optional[str] = Field( + None, description='Text prompt for the generation', max_length=1000 + ) + ratio: RunwayAspectRatioEnum + seed: int = Field( + ..., description='Random seed for generation', ge=0, le=4294967295 + ) + + +class TextResponseFormatConfiguration( + RootModel[ + Union[ + ResponseFormatText, TextResponseFormatJsonSchema, ResponseFormatJsonObject + ] + ] +): + root: Union[ + ResponseFormatText, TextResponseFormatJsonSchema, ResponseFormatJsonObject + ] = Field( + ..., + description='An object specifying the format that the model must output.\n\nConfiguring `{ "type": "json_schema" }` enables Structured Outputs, \nwhich ensures the model will match your supplied JSON schema. Learn more in the \n[Structured Outputs guide](/docs/guides/structured-outputs).\n\nThe default format is `{ "type": "text" }` with no additional options.\n\n**Not recommended for gpt-4o and newer models:**\n\nSetting to `{ "type": "json_object" }` enables the older JSON mode, which\nensures the message the model generates is valid JSON. Using `json_schema`\nis preferred for models that support it.\n', + ) + + +class Tool( + RootModel[ + Union[ + FileSearchTool, FunctionTool, WebSearchPreviewTool, ComputerUsePreviewTool + ] + ] +): + root: Union[ + FileSearchTool, FunctionTool, WebSearchPreviewTool, ComputerUsePreviewTool + ] = Field(..., discriminator='type') + + +class EasyInputMessage(BaseModel): + content: Union[str, InputMessageContentList] = Field( + ..., + description='Text, image, or audio input to the model, used to generate a response.\nCan also contain previous assistant responses.\n', + ) + role: Role = Field( + ..., + description='The role of the message input. One of `user`, `assistant`, `system`, or\n`developer`.\n', + ) + type: Optional[Type2] = Field( + None, description='The type of the message input. Always `message`.\n' + ) + + +class GeminiCandidate(BaseModel): + citationMetadata: Optional[GeminiCitationMetadata] = None + content: Optional[GeminiContent] = None + finishReason: Optional[str] = None + safetyRatings: Optional[List[GeminiSafetyRating]] = None + + +class GeminiGenerateContentResponse(BaseModel): + candidates: Optional[List[GeminiCandidate]] = None + promptFeedback: Optional[GeminiPromptFeedback] = None + + +class ImagenGenerateImageRequest(BaseModel): + instances: List[ImagenImageGenerationInstance] + parameters: ImagenImageGenerationParameters + + +class InputMessage(BaseModel): + content: Optional[InputMessageContentList] = None + role: Optional[Role1] = None + status: Optional[Status2] = None + type: Optional[Type9] = None + + +class Item( + RootModel[ + Union[ + InputMessage, + OutputMessage, + FileSearchToolCall, + ComputerToolCall, + WebSearchToolCall, + FunctionToolCall, + ReasoningItem, + ] + ] +): + root: Union[ + InputMessage, + OutputMessage, + FileSearchToolCall, + ComputerToolCall, + WebSearchToolCall, + FunctionToolCall, + ReasoningItem, + ] = Field(..., description='Content item used to generate a response.\n') class StripeCharge(BaseModel): @@ -4914,16 +5026,15 @@ class StripePaymentIntent(BaseModel): class LumaGeneration(BaseModel): - id: Optional[UUID] = Field(None, description='The ID of the generation') - generation_type: Optional[LumaGenerationType] = None - state: Optional[LumaState] = None - failure_reason: Optional[str] = Field( - None, description='The reason for the state of the generation' - ) + assets: Optional[LumaAssets] = None created_at: Optional[datetime] = Field( None, description='The date and time when the generation was created' ) - assets: Optional[LumaAssets] = None + failure_reason: Optional[str] = Field( + None, description='The reason for the state of the generation' + ) + generation_type: Optional[LumaGenerationType] = None + id: Optional[UUID] = Field(None, description='The ID of the generation') model: Optional[str] = Field(None, description='The model used for the generation') request: Optional[ Union[ @@ -4933,237 +5044,117 @@ class LumaGeneration(BaseModel): LumaAudioGenerationRequest, ] ] = Field(None, description='The request of the generation') + state: Optional[LumaState] = None -class RunwayImageToVideoRequest(BaseModel): - promptImage: RunwayPromptImageObject - seed: int = Field( - ..., description='Random seed for generation', ge=0, le=4294967295 +class OutputItem( + RootModel[ + Union[ + OutputMessage, + FileSearchToolCall, + FunctionToolCall, + WebSearchToolCall, + ComputerToolCall, + ReasoningItem, + ] + ] +): + root: Union[ + OutputMessage, + FileSearchToolCall, + FunctionToolCall, + WebSearchToolCall, + ComputerToolCall, + ReasoningItem, + ] + + +class Text(BaseModel): + format: Optional[TextResponseFormatConfiguration] = None + + +class ResponseProperties(BaseModel): + instructions: Optional[str] = Field( + None, + description="Inserts a system (or developer) message as the first item in the model's context.\n\nWhen using along with `previous_response_id`, the instructions from a previous\nresponse will not be carried over to the next response. This makes it simple\nto swap out system (or developer) messages in new responses.\n", ) - model: RunwayModelEnum = Field(..., description='Model to use for generation') - promptText: Optional[str] = Field( - None, description='Text prompt for the generation', max_length=1000 + max_output_tokens: Optional[int] = Field( + None, + description='An upper bound for the number of tokens that can be generated for a response, including visible output tokens and [reasoning tokens](/docs/guides/reasoning).\n', ) - duration: RunwayDurationEnum = Field( - ..., description='The number of seconds of duration for the output video.' + model: Optional[OpenAIModels] = None + previous_response_id: Optional[str] = Field( + None, + description='The unique ID of the previous response to the model. Use this to\ncreate multi-turn conversations. Learn more about \n[conversation state](/docs/guides/conversation-state).\n', ) - ratio: RunwayAspectRatioEnum = Field( + reasoning: Optional[Reasoning] = None + text: Optional[Text] = None + tool_choice: Optional[ + Union[ToolChoiceOptions, ToolChoiceTypes, ToolChoiceFunction] + ] = Field( + None, + description='How the model should select which tool (or tools) to use when generating\na response. See the `tools` parameter to see how to specify which tools\nthe model can call.\n', + ) + tools: Optional[List[Tool]] = None + truncation: Optional[Truncation1] = Field( + 'disabled', + description="The truncation strategy to use for the model response.\n- `auto`: If the context of this response and previous ones exceeds\n the model's context window size, the model will truncate the \n response to fit the context window by dropping input items in the\n middle of the conversation. \n- `disabled` (default): If a model response will exceed the context window \n size for a model, the request will fail with a 400 error.\n", + ) + + +class InputItem(RootModel[Union[EasyInputMessage, Item]]): + root: Union[EasyInputMessage, Item] + + +class OpenAICreateResponse(CreateModelResponseProperties, ResponseProperties): + include: Optional[List[Includable]] = Field( + None, + description='Specify additional output data to include in the model response. Currently\nsupported values are:\n- `file_search_call.results`: Include the search results of\n the file search tool call.\n- `message.input_image.image_url`: Include image urls from the input message.\n- `computer_call_output.output.image_url`: Include image urls from the computer call output.\n', + ) + input: Union[str, List[InputItem]] = Field( ..., - description='The resolution (aspect ratio) of the output video. Allowable values depend on the selected model. 1280:768 and 768:1280 are only supported for gen3a_turbo.', + description='Text, image, or file inputs to the model, used to generate a response.\n\nLearn more:\n- [Text inputs and outputs](/docs/guides/text)\n- [Image inputs](/docs/guides/images)\n- [File inputs](/docs/guides/pdf-files)\n- [Conversation state](/docs/guides/conversation-state)\n- [Function calling](/docs/guides/function-calling)\n', ) - - -class RunwayTaskStatusResponse(BaseModel): - id: Optional[str] = Field(None, description='Task ID') - status: Optional[RunwayTaskStatusEnum] = Field(None, description='Task status') - createdAt: Optional[datetime] = Field(None, description='Task creation timestamp') - output: Optional[List[str]] = Field(None, description='Array of output video URLs') - - -class PikaHTTPValidationError(BaseModel): - detail: Optional[List[PikaValidationError]] = Field(None, title='Detail') - - -class PikaBodyGenerate22T2vGenerate22T2vPost(BaseModel): - promptText: str = Field(..., title='Prompttext') - negativePrompt: Optional[str] = Field(None, title='Negativeprompt') - seed: Optional[int] = Field(None, title='Seed') - resolution: Optional[PikaResolutionEnum] = Field('1080p', title='Resolution') - duration: Optional[PikaDurationEnum] = Field(5, title='Duration') - aspectRatio: Optional[float] = Field( - 1.7777777777777777, - description='Aspect ratio (width / height)', - ge=0.4, - le=2.5, - title='Aspectratio', + parallel_tool_calls: Optional[bool] = Field( + True, description='Whether to allow the model to run tool calls in parallel.\n' ) - - -class PikaBodyGenerate22I2vGenerate22I2vPost(BaseModel): - image: Optional[StrictBytes] = Field(None, title='Image') - promptText: Optional[str] = Field(None, title='Prompttext') - negativePrompt: Optional[str] = Field(None, title='Negativeprompt') - seed: Optional[int] = Field(None, title='Seed') - resolution: Optional[PikaResolutionEnum] = Field('1080p', title='Resolution') - duration: Optional[PikaDurationEnum] = Field(5, title='Duration') - - -class PikaBodyGenerate22KeyframeGenerate22PikaframesPost(BaseModel): - keyFrames: Optional[List[StrictBytes]] = Field( - None, description='Array of keyframe images', title='Keyframes' + store: Optional[bool] = Field( + True, + description='Whether to store the generated model response for later retrieval via\nAPI.\n', ) - promptText: str = Field(..., title='Prompttext') - negativePrompt: Optional[str] = Field(None, title='Negativeprompt') - seed: Optional[int] = Field(None, title='Seed') - resolution: Optional[PikaResolutionEnum] = Field('1080p', title='Resolution') - duration: Optional[int] = Field(None, ge=5, le=10, title='Duration') + stream: Optional[bool] = Field( + False, + description='If set to true, the model response data will be streamed to the client\nas it is generated using [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format).\nSee the [Streaming section below](/docs/api-reference/responses-streaming)\nfor more information.\n', + ) + usage: Optional[ResponseUsage] = None -class PikaVideoResponse(BaseModel): - id: str = Field(..., title='Id') - status: PikaStatusEnum = Field( - ..., description='The status of the video', title='Status' - ) - url: Optional[str] = Field(None, title='Url') - progress: Optional[int] = Field(None, title='Progress') - - -class Node(BaseModel): - id: Optional[str] = Field(None, description='The unique identifier of the node.') - name: Optional[str] = Field(None, description='The display name of the node.') - category: Optional[str] = Field(None, description='The category of the node.') - description: Optional[str] = None - author: Optional[str] = None - license: Optional[str] = Field( - None, description="The path to the LICENSE file in the node's repository." - ) - icon: Optional[str] = Field(None, description="URL to the node's icon.") - repository: Optional[str] = Field(None, description="URL to the node's repository.") - tags: Optional[List[str]] = None - latest_version: Optional[NodeVersion] = Field( - None, description='The latest version of the node.' - ) - rating: Optional[float] = Field(None, description='The average rating of the node.') - downloads: Optional[int] = Field( - None, description='The number of downloads of the node.' - ) - publisher: Optional[Publisher] = Field( - None, description='The publisher of the node.' - ) - status: Optional[NodeStatus] = Field(None, description='The status of the node.') - status_detail: Optional[str] = Field( - None, description='The status detail of the node.' - ) - translations: Optional[Dict[str, Dict[str, Any]]] = None - - -class KlingVideoEffectsRequest(BaseModel): - effect_scene: Union[KlingDualCharacterEffectsScene, KlingSingleImageEffectsScene] - input: KlingVideoEffectsInput - callback_url: Optional[AnyUrl] = Field( +class OpenAIResponse(ModelResponseProperties, ResponseProperties): + created_at: Optional[float] = Field( None, - description='The callback notification address for the result of this task.', + description='Unix timestamp (in seconds) of when this Response was created.', ) - external_task_id: Optional[str] = Field( + error: Optional[ResponseError] = None + id: Optional[str] = Field(None, description='Unique identifier for this Response.') + incomplete_details: Optional[IncompleteDetails] = Field( + None, description='Details about why the response is incomplete.\n' + ) + object: Optional[Object] = Field( + None, description='The object type of this resource - always set to `response`.' + ) + output: Optional[List[OutputItem]] = Field( None, - description='Customized Task ID. Must be unique within a single user account.', + description="An array of content items generated by the model.\n\n- The length and order of items in the `output` array is dependent\n on the model's response.\n- Rather than accessing the first item in the `output` array and \n assuming it's an `assistant` message with the content generated by\n the model, you might consider using the `output_text` property where\n supported in SDKs.\n", ) - - -class StripeCharge(BaseModel): - id: Optional[str] = None - object: Optional[Object2] = None - amount: Optional[int] = None - amount_captured: Optional[int] = None - amount_refunded: Optional[int] = None - application: Optional[str] = None - application_fee: Optional[str] = None - application_fee_amount: Optional[int] = None - balance_transaction: Optional[str] = None - billing_details: Optional[StripeBillingDetails] = None - calculated_statement_descriptor: Optional[str] = None - captured: Optional[bool] = None - created: Optional[int] = None - currency: Optional[str] = None - customer: Optional[str] = None - description: Optional[str] = None - destination: Optional[Any] = None - dispute: Optional[Any] = None - disputed: Optional[bool] = None - failure_balance_transaction: Optional[Any] = None - failure_code: Optional[Any] = None - failure_message: Optional[Any] = None - fraud_details: Optional[Dict[str, Any]] = None - invoice: Optional[Any] = None - livemode: Optional[bool] = None - metadata: Optional[Dict[str, Any]] = None - on_behalf_of: Optional[Any] = None - order: Optional[Any] = None - outcome: Optional[StripeOutcome] = None - paid: Optional[bool] = None - payment_intent: Optional[str] = None - payment_method: Optional[str] = None - payment_method_details: Optional[StripePaymentMethodDetails] = None - radar_options: Optional[Dict[str, Any]] = None - receipt_email: Optional[str] = None - receipt_number: Optional[str] = None - receipt_url: Optional[str] = None - refunded: Optional[bool] = None - refunds: Optional[StripeRefundList] = None - review: Optional[Any] = None - shipping: Optional[StripeShipping] = None - source: Optional[Any] = None - source_transfer: Optional[Any] = None - statement_descriptor: Optional[Any] = None - statement_descriptor_suffix: Optional[Any] = None - status: Optional[str] = None - transfer_data: Optional[Any] = None - transfer_group: Optional[Any] = None - - -class StripeChargeList(BaseModel): - object: Optional[str] = None - data: Optional[List[StripeCharge]] = None - has_more: Optional[bool] = None - total_count: Optional[int] = None - url: Optional[str] = None - - -class StripePaymentIntent(BaseModel): - id: Optional[str] = None - object: Optional[Object1] = None - amount: Optional[int] = None - amount_capturable: Optional[int] = None - amount_details: Optional[StripeAmountDetails] = None - amount_received: Optional[int] = None - application: Optional[str] = None - application_fee_amount: Optional[int] = None - automatic_payment_methods: Optional[Any] = None - canceled_at: Optional[int] = None - cancellation_reason: Optional[str] = None - capture_method: Optional[str] = None - charges: Optional[StripeChargeList] = None - client_secret: Optional[str] = None - confirmation_method: Optional[str] = None - created: Optional[int] = None - currency: Optional[str] = None - customer: Optional[str] = None - description: Optional[str] = None - invoice: Optional[str] = None - last_payment_error: Optional[Any] = None - latest_charge: Optional[str] = None - livemode: Optional[bool] = None - metadata: Optional[Dict[str, Any]] = None - next_action: Optional[Any] = None - on_behalf_of: Optional[Any] = None - payment_method: Optional[str] = None - payment_method_configuration_details: Optional[Any] = None - payment_method_options: Optional[StripePaymentMethodOptions] = None - payment_method_types: Optional[List[str]] = None - processing: Optional[Any] = None - receipt_email: Optional[str] = None - review: Optional[Any] = None - setup_future_usage: Optional[Any] = None - shipping: Optional[StripeShipping] = None - source: Optional[Any] = None - statement_descriptor: Optional[Any] = None - statement_descriptor_suffix: Optional[Any] = None - status: Optional[str] = None - transfer_data: Optional[Any] = None - transfer_group: Optional[Any] = None - - -class Data8(BaseModel): - object: Optional[StripePaymentIntent] = None - - -class StripeEvent(BaseModel): - id: str - object: Object - api_version: Optional[str] = None - created: Optional[int] = None - data: Data8 - livemode: Optional[bool] = None - pending_webhooks: Optional[int] = None - request: Optional[StripeRequestInfo] = None - type: Type + output_text: Optional[str] = Field( + None, + description='SDK-only convenience property that contains the aggregated text output \nfrom all `output_text` items in the `output` array, if any are present. \nSupported in the Python and JavaScript SDKs.\n', + ) + parallel_tool_calls: Optional[bool] = Field( + True, description='Whether to allow the model to run tool calls in parallel.\n' + ) + status: Optional[Status6] = Field( + None, + description='The status of the response generation. One of `completed`, `failed`, `in_progress`, or `incomplete`.', + ) + usage: Optional[ResponseUsage] = None diff --git a/comfy_api_nodes/nodes_openai.py b/comfy_api_nodes/nodes_openai.py index ce8054afc..85cd3cf5a 100644 --- a/comfy_api_nodes/nodes_openai.py +++ b/comfy_api_nodes/nodes_openai.py @@ -1,29 +1,86 @@ import io +from typing import TypedDict, Optional +import json +import os +import time +import re +import uuid +from enum import Enum from inspect import cleandoc import numpy as np import torch from PIL import Image - from comfy.comfy_types.node_typing import IO, ComfyNodeABC, InputTypeDict +from server import PromptServer +import folder_paths from comfy_api_nodes.apis import ( OpenAIImageGenerationRequest, OpenAIImageEditRequest, OpenAIImageGenerationResponse, + OpenAICreateResponse, + OpenAIResponse, + CreateModelResponseProperties, + Item, + Includable, + OutputContent, + InputImageContent, + Detail, + InputTextContent, + InputMessage, + InputMessageContentList, + InputContent, + InputFileContent, ) from comfy_api_nodes.apis.client import ( ApiEndpoint, HttpMethod, SynchronousOperation, + PollingOperation, + EmptyRequest, ) from comfy_api_nodes.apinode_utils import ( downscale_image_tensor, validate_and_cast_response, validate_string, + tensor_to_base64_string, + text_filepath_to_data_uri, ) +from comfy_api_nodes.mapper_utils import model_field_to_node_input + + +RESPONSES_ENDPOINT = "/proxy/openai/v1/responses" +STARTING_POINT_ID_PATTERN = r"" + + +class HistoryEntry(TypedDict): + """Type definition for a single history entry in the chat.""" + + prompt: str + response: str + response_id: str + timestamp: float + + +class ChatHistory(TypedDict): + """Type definition for the chat history dictionary.""" + + __annotations__: dict[str, list[HistoryEntry]] + + +class SupportedOpenAIModel(str, Enum): + o4_mini = "o4-mini" + o1 = "o1" + o3 = "o3" + o1_pro = "o1-pro" + gpt_4o = "gpt-4o" + gpt_4_1 = "gpt-4.1" + gpt_4_1_mini = "gpt-4.1-mini" + gpt_4_1_nano = "gpt-4.1-nano" + class OpenAIDalle2(ComfyNodeABC): """ @@ -115,7 +172,7 @@ class OpenAIDalle2(ComfyNodeABC): n=1, size="1024x1024", unique_id=None, - **kwargs + **kwargs, ): validate_string(prompt, strip_whitespace=False) model = "dall-e-2" @@ -262,7 +319,7 @@ class OpenAIDalle3(ComfyNodeABC): quality="standard", size="1024x1024", unique_id=None, - **kwargs + **kwargs, ): validate_string(prompt, strip_whitespace=False) model = "dall-e-3" @@ -400,12 +457,12 @@ class OpenAIGPTImage1(ComfyNodeABC): n=1, size="1024x1024", unique_id=None, - **kwargs + **kwargs, ): validate_string(prompt, strip_whitespace=False) model = "gpt-image-1" path = "/proxy/openai/images/generations" - content_type="application/json" + content_type = "application/json" request_class = OpenAIImageGenerationRequest img_binaries = [] mask_binary = None @@ -414,7 +471,7 @@ class OpenAIGPTImage1(ComfyNodeABC): if image is not None: path = "/proxy/openai/images/edits" request_class = OpenAIImageEditRequest - content_type ="multipart/form-data" + content_type = "multipart/form-data" batch_size = image.shape[0] @@ -486,17 +543,466 @@ class OpenAIGPTImage1(ComfyNodeABC): return (img_tensor,) -# A dictionary that contains all nodes you want to export with their names -# NOTE: names should be globally unique +class OpenAITextNode(ComfyNodeABC): + """ + Base class for OpenAI text generation nodes. + """ + + RETURN_TYPES = (IO.STRING,) + FUNCTION = "api_call" + CATEGORY = "api node/text/OpenAI" + API_NODE = True + + +class OpenAIChatNode(OpenAITextNode): + """ + Node to generate text responses from an OpenAI model. + """ + + def __init__(self) -> None: + """Initialize the chat node with a new session ID and empty history.""" + self.current_session_id: str = str(uuid.uuid4()) + self.history: dict[str, list[HistoryEntry]] = {} + self.previous_response_id: Optional[str] = None + + @classmethod + def INPUT_TYPES(cls) -> InputTypeDict: + return { + "required": { + "prompt": ( + IO.STRING, + { + "multiline": True, + "default": "", + "tooltip": "Text inputs to the model, used to generate a response.", + }, + ), + "persist_context": ( + IO.BOOLEAN, + { + "default": True, + "tooltip": "Persist chat context between calls (multi-turn conversation)", + }, + ), + "model": model_field_to_node_input( + IO.COMBO, + OpenAICreateResponse, + "model", + enum_type=SupportedOpenAIModel, + ), + }, + "optional": { + "images": ( + IO.IMAGE, + { + "default": None, + "tooltip": "Optional image(s) to use as context for the model. To include multiple images, you can use the Batch Images node.", + }, + ), + "files": ( + "OPENAI_INPUT_FILES", + { + "default": None, + "tooltip": "Optional file(s) to use as context for the model. Accepts inputs from the OpenAI Chat Input Files node.", + }, + ), + "advanced_options": ( + "OPENAI_CHAT_CONFIG", + { + "default": None, + "tooltip": "Optional configuration for the model. Accepts inputs from the OpenAI Chat Advanced Options node.", + }, + ), + }, + "hidden": { + "auth_token": "AUTH_TOKEN_COMFY_ORG", + "comfy_api_key": "API_KEY_COMFY_ORG", + "unique_id": "UNIQUE_ID", + }, + } + + DESCRIPTION = "Generate text responses from an OpenAI model." + + def get_result_response( + self, + response_id: str, + include: Optional[list[Includable]] = None, + auth_kwargs: Optional[dict[str, str]] = None, + ) -> OpenAIResponse: + """ + Retrieve a model response with the given ID from the OpenAI API. + + Args: + response_id (str): The ID of the response to retrieve. + include (Optional[List[Includable]]): Additional fields to include + in the response. See the `include` parameter for Response + creation above for more information. + + """ + return PollingOperation( + poll_endpoint=ApiEndpoint( + path=f"{RESPONSES_ENDPOINT}/{response_id}", + method=HttpMethod.GET, + request_model=EmptyRequest, + response_model=OpenAIResponse, + query_params={"include": include}, + ), + completed_statuses=["completed"], + failed_statuses=["failed"], + status_extractor=lambda response: response.status, + auth_kwargs=auth_kwargs, + ).execute() + + def get_message_content_from_response( + self, response: OpenAIResponse + ) -> list[OutputContent]: + """Extract message content from the API response.""" + for output in response.output: + if output.root.type == "message": + return output.root.content + raise TypeError("No output message found in response") + + def get_text_from_message_content( + self, message_content: list[OutputContent] + ) -> str: + """Extract text content from message content.""" + for content_item in message_content: + if content_item.root.type == "output_text": + return str(content_item.root.text) + return "No text output found in response" + + def get_history_text(self, session_id: str) -> str: + """Convert the entire history for a given session to JSON string.""" + return json.dumps(self.history[session_id]) + + def display_history_on_node(self, session_id: str, node_id: str) -> None: + """Display formatted chat history on the node UI.""" + render_spec = { + "node_id": node_id, + "component": "ChatHistoryWidget", + "props": { + "history": self.get_history_text(session_id), + }, + } + PromptServer.instance.send_sync( + "display_component", + render_spec, + ) + + def add_to_history( + self, session_id: str, prompt: str, output_text: str, response_id: str + ) -> None: + """Add a new entry to the chat history.""" + if session_id not in self.history: + self.history[session_id] = [] + self.history[session_id].append( + { + "prompt": prompt, + "response": output_text, + "response_id": response_id, + "timestamp": time.time(), + } + ) + + def parse_output_text_from_response(self, response: OpenAIResponse) -> str: + """Extract text output from the API response.""" + message_contents = self.get_message_content_from_response(response) + return self.get_text_from_message_content(message_contents) + + def generate_new_session_id(self) -> str: + """Generate a new unique session ID.""" + return str(uuid.uuid4()) + + def get_session_id(self, persist_context: bool) -> str: + """Get the current or generate a new session ID based on context persistence.""" + return ( + self.current_session_id + if persist_context + else self.generate_new_session_id() + ) + + def tensor_to_input_image_content( + self, image: torch.Tensor, detail_level: Detail = "auto" + ) -> InputImageContent: + """Convert a tensor to an input image content object.""" + return InputImageContent( + detail=detail_level, + image_url=f"data:image/png;base64,{tensor_to_base64_string(image)}", + type="input_image", + ) + + def create_input_message_contents( + self, + prompt: str, + image: Optional[torch.Tensor] = None, + files: Optional[list[InputFileContent]] = None, + ) -> InputMessageContentList: + """Create a list of input message contents from prompt and optional image.""" + content_list: list[InputContent] = [ + InputTextContent(text=prompt, type="input_text"), + ] + if image is not None: + for i in range(image.shape[0]): + content_list.append( + self.tensor_to_input_image_content(image[i].unsqueeze(0)) + ) + if files is not None: + content_list.extend(files) + + return InputMessageContentList( + root=content_list, + ) + + def parse_response_id_from_prompt(self, prompt: str) -> Optional[str]: + """Extract response ID from prompt if it exists.""" + parsed_id = re.search(STARTING_POINT_ID_PATTERN, prompt) + return parsed_id.group(1) if parsed_id else None + + def strip_response_tag_from_prompt(self, prompt: str) -> str: + """Remove the response ID tag from the prompt.""" + return re.sub(STARTING_POINT_ID_PATTERN, "", prompt.strip()) + + def delete_history_after_response_id( + self, new_start_id: str, session_id: str + ) -> None: + """Delete history entries after a specific response ID.""" + if session_id not in self.history: + return + + new_history = [] + i = 0 + while ( + i < len(self.history[session_id]) + and self.history[session_id][i]["response_id"] != new_start_id + ): + new_history.append(self.history[session_id][i]) + i += 1 + + # Since it's the new starting point (not the response being edited), we include it as well + if i < len(self.history[session_id]): + new_history.append(self.history[session_id][i]) + + self.history[session_id] = new_history + + def api_call( + self, + prompt: str, + persist_context: bool, + model: SupportedOpenAIModel, + unique_id: Optional[str] = None, + images: Optional[torch.Tensor] = None, + files: Optional[list[InputFileContent]] = None, + advanced_options: Optional[CreateModelResponseProperties] = None, + **kwargs, + ) -> tuple[str]: + # Validate inputs + validate_string(prompt, strip_whitespace=False) + + session_id = self.get_session_id(persist_context) + response_id_override = self.parse_response_id_from_prompt(prompt) + if response_id_override: + is_starting_from_beginning = response_id_override == "start" + if is_starting_from_beginning: + self.history[session_id] = [] + previous_response_id = None + else: + previous_response_id = response_id_override + self.delete_history_after_response_id(response_id_override, session_id) + prompt = self.strip_response_tag_from_prompt(prompt) + elif persist_context: + previous_response_id = self.previous_response_id + else: + previous_response_id = None + + # Create response + create_response = SynchronousOperation( + endpoint=ApiEndpoint( + path=RESPONSES_ENDPOINT, + method=HttpMethod.POST, + request_model=OpenAICreateResponse, + response_model=OpenAIResponse, + ), + request=OpenAICreateResponse( + input=[ + Item( + root=InputMessage( + content=self.create_input_message_contents( + prompt, images, files + ), + role="user", + ) + ), + ], + store=True, + stream=False, + model=model, + previous_response_id=previous_response_id, + **( + advanced_options.model_dump(exclude_none=True) + if advanced_options + else {} + ), + ), + auth_kwargs=kwargs, + ).execute() + response_id = create_response.id + + # Get result output + result_response = self.get_result_response(response_id, auth_kwargs=kwargs) + output_text = self.parse_output_text_from_response(result_response) + + # Update history + self.add_to_history(session_id, prompt, output_text, response_id) + self.display_history_on_node(session_id, unique_id) + self.previous_response_id = response_id + + return (output_text,) + + +class OpenAIInputFiles(ComfyNodeABC): + """ + Loads and formats input files for OpenAI API. + """ + + @classmethod + def INPUT_TYPES(cls) -> InputTypeDict: + """ + For details about the supported file input types, see: + https://platform.openai.com/docs/guides/pdf-files?api-mode=responses + """ + input_dir = folder_paths.get_input_directory() + input_files = [ + f + for f in os.scandir(input_dir) + if f.is_file() + and (f.name.endswith(".txt") or f.name.endswith(".pdf")) + and f.stat().st_size < 32 * 1024 * 1024 + ] + input_files = sorted(input_files, key=lambda x: x.name) + input_files = [f.name for f in input_files] + return { + "required": { + "file": ( + IO.COMBO, + { + "tooltip": "Input files to include as context for the model.", + "options": input_files, + "default": input_files[0] if input_files else None, + }, + ), + }, + "optional": { + "OPENAI_INPUT_FILES": ( + "OPENAI_INPUT_FILES", + { + "tooltip": "An optional additional file(s) to batch together with the file loaded from this node. Allows chaining of input files so that a single message can include multiple input files.", + "default": None, + }, + ), + }, + } + + DESCRIPTION = "Loads and prepares input files (text, pdf, etc.) to include as inputs for the OpenAI Chat Node. The files will be read by the OpenAI model when generating a response. 🛈 TIP: Can be chained together with other OpenAI Input File nodes." + RETURN_TYPES = ("OPENAI_INPUT_FILES",) + FUNCTION = "prepare_files" + CATEGORY = "api node/text/OpenAI" + + def create_input_file_content(self, file_path: str) -> InputFileContent: + return InputFileContent( + file_data=text_filepath_to_data_uri(file_path), + filename=os.path.basename(file_path), + type="input_file", + ) + + def prepare_files( + self, file: str, OPENAI_INPUT_FILES: list[InputFileContent] = [] + ) -> tuple[list[InputFileContent]]: + """ + Loads and formats input files for OpenAI API. + """ + file_path = folder_paths.get_annotated_filepath(file) + input_file_content = self.create_input_file_content(file_path) + files = [input_file_content] + OPENAI_INPUT_FILES + return (files,) + + +class OpenAIChatConfig(ComfyNodeABC): + """Allows setting additional configuration for the OpenAI Chat Node.""" + + RETURN_TYPES = ("OPENAI_CHAT_CONFIG",) + FUNCTION = "configure" + DESCRIPTION = ( + "Allows specifying advanced configuration options for the OpenAI Chat Nodes." + ) + CATEGORY = "api node/text/OpenAI" + + @classmethod + def INPUT_TYPES(cls) -> InputTypeDict: + return { + "required": { + "truncation": ( + IO.COMBO, + { + "options": ["auto", "disabled"], + "default": "auto", + "tooltip": "The truncation strategy to use for the model response. auto: If the context of this response and previous ones exceeds the model's context window size, the model will truncate the response to fit the context window by dropping input items in the middle of the conversation.disabled: If a model response will exceed the context window size for a model, the request will fail with a 400 error", + }, + ), + }, + "optional": { + "max_output_tokens": model_field_to_node_input( + IO.INT, + OpenAICreateResponse, + "max_output_tokens", + min=16, + default=4096, + max=16384, + tooltip="An upper bound for the number of tokens that can be generated for a response, including visible output tokens", + ), + "instructions": model_field_to_node_input( + IO.STRING, OpenAICreateResponse, "instructions", multiline=True + ), + }, + } + + def configure( + self, + truncation: bool, + instructions: Optional[str] = None, + max_output_tokens: Optional[int] = None, + ) -> tuple[CreateModelResponseProperties]: + """ + Configure advanced options for the OpenAI Chat Node. + + Note: + While `top_p` and `temperature` are listed as properties in the + spec, they are not supported for all models (e.g., o4-mini). + They are not exposed as inputs at all to avoid having to manually + remove depending on model choice. + """ + return ( + CreateModelResponseProperties( + instructions=instructions, + truncation=truncation, + max_output_tokens=max_output_tokens, + ), + ) + + NODE_CLASS_MAPPINGS = { "OpenAIDalle2": OpenAIDalle2, "OpenAIDalle3": OpenAIDalle3, "OpenAIGPTImage1": OpenAIGPTImage1, + "OpenAIChatNode": OpenAIChatNode, + "OpenAIInputFiles": OpenAIInputFiles, + "OpenAIChatConfig": OpenAIChatConfig, } -# A dictionary that contains the friendly/humanly readable titles for the nodes NODE_DISPLAY_NAME_MAPPINGS = { "OpenAIDalle2": "OpenAI DALL·E 2", "OpenAIDalle3": "OpenAI DALL·E 3", "OpenAIGPTImage1": "OpenAI GPT Image 1", + "OpenAIChatNode": "OpenAI Chat", + "OpenAIInputFiles": "OpenAI Chat Input Files", + "OpenAIChatConfig": "OpenAI Chat Advanced Options", }