chore: sync from ark-apis 134e24deb3

VERSION

@@ -1 +1 @@
-0.12.0
+0.13.0

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "arkruntime"
-version = "0.12.0"
+version = "0.13.0"
 description = "Ark runtime SDK for Python"
 readme = "README.md"
 requires-python = ">=3.8"

src/arkruntime/types/audio_generation/__init__.py

@@ -0,0 +1,90 @@
+# Generated by datamodel-code-generator from ark-apis typespec/openapi.
+# DO NOT EDIT — regenerate with `make gen-py-<api>` in ark-apis.
+#
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from .audio_config import AudioConfig
+from .audio_content import AudioContent
+from .audio_generation_request import AudioGenerationRequest
+from .audio_generation_response import AudioGenerationResponse
+from .audio_role import AudioRole
+from .audio_url import AudioURL
+from .content import Content
+from .dubbing_config import DubbingConfig
+from .effective_dubbing_config import EffectiveDubbingConfig
+from .error_response import ErrorResponse
+from .glossaries import Glossaries
+from .glossary import Glossary
+from .image_content import ImageContent
+from .image_url import ImageURL
+from .include import Include
+from .non_empty_string import NonEmptyString
+from .output_format import OutputFormat
+from .pitch_rate import PitchRate
+from .rate import Rate
+from .response_audio import ResponseAudio
+from .response_content import ResponseContent
+from .response_error import ResponseError
+from .response_subtitle import ResponseSubtitle
+from .sample_rate import SampleRate
+from .target_language import TargetLanguage
+from .text_content import TextContent
+from .usage import Usage
+from .video_content import VideoContent
+from .video_role import VideoRole
+from .video_url import VideoURL
+
+__all__ = [
+    "AudioConfig",
+    "AudioContent",
+    "AudioGenerationRequest",
+    "AudioGenerationResponse",
+    "AudioRole",
+    "AudioURL",
+    "Content",
+    "DubbingConfig",
+    "EffectiveDubbingConfig",
+    "ErrorResponse",
+    "Glossaries",
+    "Glossary",
+    "ImageContent",
+    "ImageURL",
+    "Include",
+    "NonEmptyString",
+    "OutputFormat",
+    "PitchRate",
+    "Rate",
+    "ResponseAudio",
+    "ResponseContent",
+    "ResponseError",
+    "ResponseSubtitle",
+    "SampleRate",
+    "TargetLanguage",
+    "TextContent",
+    "Usage",
+    "VideoContent",
+    "VideoRole",
+    "VideoURL",
+]
+from .audio_config_param import AudioConfigParam
+from .audio_content_param import AudioContentParam
+from .audio_generation_request_param import AudioGenerationRequestParam
+from .audio_url_param import AudioURLParam
+from .content_param import ContentParam
+from .dubbing_config_param import DubbingConfigParam
+from .glossaries_param import GlossariesParam
+from .glossary_param import GlossaryParam
+from .image_content_param import ImageContentParam
+from .image_url_param import ImageURLParam
+from .non_empty_string_param import NonEmptyStringParam
+from .pitch_rate_param import PitchRateParam
+from .rate_param import RateParam
+from .text_content_param import TextContentParam
+from .video_content_param import VideoContentParam
+from .video_url_param import VideoURLParam
+
+# Hand-written extras (preserved across regen via Makefile rsync --exclude=*_shim.py).
+from ._init_extras_shim import *  # noqa: F401,F403

src/arkruntime/types/audio_generation/_init_extras_shim.py

@@ -0,0 +1,12 @@
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+"""Hand-written extras shim for the audio_generation API (empty stub).
+
+Preserved across regen via the Makefile rsync --exclude=*_shim.py rule.
+Add re-exports here as needed.
+"""
+
+from __future__ import annotations
+
+__all__: list[str] = []

src/arkruntime/types/audio_generation/audio_config.py

@@ -0,0 +1,34 @@
+# Generated by datamodel-code-generator from ark-apis typespec/openapi.
+# DO NOT EDIT — regenerate with `make gen-py-<api>` in ark-apis.
+#
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from typing import Optional
+
+from arkruntime._models import BaseModel
+
+from .pitch_rate import PitchRate
+from .rate import Rate
+from .sample_rate import SampleRate
+
+
+class AudioConfig(BaseModel):
+    sample_rate: Optional[SampleRate] = None
+    """
+    Defaults to 44100 for wav/mp3/pcm and 48000 for ogg_opus. ogg_opus only accepts 48000. Explicit zero is invalid.
+    """
+    speech_rate: Optional[Rate] = None
+    """
+    Reference generation only. Default 0; -50 means 0.5x and 100 means 2x. Must be omitted (or null) for dubbing/separation, even when the value would be 0.
+    """
+    loudness_rate: Optional[Rate] = None
+    """
+    Reference generation only. Default 0; -50 means 0.5x and 100 means 2x. Must be omitted (or null) for dubbing/separation.
+    """
+    pitch_rate: Optional[PitchRate] = None
+    """
+    Reference generation only. Default 0. Must be omitted (or null) for dubbing/separation.
+    """

src/arkruntime/types/audio_generation/audio_config_param.py

@@ -0,0 +1,29 @@
+# Generated by datamodel-code-generator from ark-apis typespec/openapi.
+# DO NOT EDIT — regenerate with `make gen-py-<api>` in ark-apis.
+#
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from typing import Optional
+
+from typing_extensions import TypedDict
+
+from .pitch_rate_param import PitchRateParam
+from .rate_param import RateParam
+from .sample_rate import SampleRate
+
+
+class AudioConfigParam(TypedDict, total=False):
+    sample_rate: Optional[Optional[SampleRate]]
+    """Defaults to 44100 for wav/mp3/pcm and 48000 for ogg_opus. ogg_opus only accepts 48000. Explicit zero is invalid."""
+
+    speech_rate: Optional[Optional[RateParam]]
+    """Reference generation only. Default 0; -50 means 0.5x and 100 means 2x. Must be omitted (or null) for dubbing/separation, even when the value would be 0."""
+
+    loudness_rate: Optional[Optional[RateParam]]
+    """Reference generation only. Default 0; -50 means 0.5x and 100 means 2x. Must be omitted (or null) for dubbing/separation."""
+
+    pitch_rate: Optional[Optional[PitchRateParam]]
+    """Reference generation only. Default 0. Must be omitted (or null) for dubbing/separation."""

src/arkruntime/types/audio_generation/audio_content.py

@@ -0,0 +1,23 @@
+# Generated by datamodel-code-generator from ark-apis typespec/openapi.
+# DO NOT EDIT — regenerate with `make gen-py-<api>` in ark-apis.
+#
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from typing import Literal
+
+from arkruntime._models import BaseModel
+
+from .audio_role import AudioRole
+from .audio_url import AudioURL
+
+
+class AudioContent(BaseModel):
+    type: Literal["audio_url"]
+    audio_url: AudioURL
+    role: AudioRole
+    """
+    Required. reference_audio selects reference generation; separate_audio selects audio separation.
+    """

src/arkruntime/types/audio_generation/audio_content_param.py

@@ -0,0 +1,23 @@
+# Generated by datamodel-code-generator from ark-apis typespec/openapi.
+# DO NOT EDIT — regenerate with `make gen-py-<api>` in ark-apis.
+#
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from typing import Literal
+
+from typing_extensions import Required, TypedDict
+
+from .audio_role import AudioRole
+from .audio_url_param import AudioURLParam
+
+
+class AudioContentParam(TypedDict, total=False):
+    type: Required[Literal["audio_url"]]
+
+    audio_url: Required[AudioURLParam]
+
+    role: Required[AudioRole]
+    """Required. reference_audio selects reference generation; separate_audio selects audio separation."""

src/arkruntime/types/audio_generation/audio_generation_request.py

@@ -0,0 +1,55 @@
+# Generated by datamodel-code-generator from ark-apis typespec/openapi.
+# DO NOT EDIT — regenerate with `make gen-py-<api>` in ark-apis.
+#
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from typing import List, Optional
+
+from typing_extensions import Annotated
+
+from arkruntime._models import BaseModel
+from pydantic import Field
+
+from .audio_config import AudioConfig
+from .content import Content
+from .dubbing_config import DubbingConfig
+from .include import Include
+from .non_empty_string import NonEmptyString
+from .output_format import OutputFormat
+
+
+class AudioGenerationRequest(BaseModel):
+    """
+    Synchronous reference generation, audio separation or video dubbing.
+    Static scene rules are attached to required content via dependentSchemas;
+    media and deployment checks run on the server.
+    """
+
+    model: NonEmptyString
+    """
+    Model ID or Endpoint ID. Whitespace-only values are rejected.
+    """
+    content: Annotated[List[Content], Field(min_length=1)]
+    dubbing_config: Optional[DubbingConfig] = None
+    """
+    Required for dubbing; forbidden in reference generation and separation.
+    """
+    output_format: Optional[OutputFormat] = "wav"
+    """
+    Output audio format. Omitted or null resolves to wav.
+    """
+    audio_config: Optional[AudioConfig] = None
+    """
+    Optional output parameters. Defaults are applied by the service.
+    """
+    include: Optional[List[Include]] = None
+    """
+    Additional artifacts; omitted/null/empty means none. video_url requires a video input. subtitles_source.* is only allowed for dubbing. Separation forbids a non-empty include.
+    """
+    watermark: Optional[bool] = False
+    """
+    Append an audible rhythm watermark at the end. Defaults to false. Only supported in the cn-beijing deployment region; overseas deployments reject true with InvalidParameter (param=watermark).
+    """

src/arkruntime/types/audio_generation/audio_generation_request_param.py

@@ -0,0 +1,46 @@
+# Generated by datamodel-code-generator from ark-apis typespec/openapi.
+# DO NOT EDIT — regenerate with `make gen-py-<api>` in ark-apis.
+#
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from typing import List, Optional
+
+from typing_extensions import Required, TypedDict
+
+from .audio_config_param import AudioConfigParam
+from .content_param import ContentParam
+from .dubbing_config_param import DubbingConfigParam
+from .include import Include
+from .non_empty_string_param import NonEmptyStringParam
+from .output_format import OutputFormat
+
+
+class AudioGenerationRequestParam(TypedDict, total=False):
+    """
+    Synchronous reference generation, audio separation or video dubbing.
+    Static scene rules are attached to required content via dependentSchemas;
+    media and deployment checks run on the server.
+    """
+
+    model: Required[NonEmptyStringParam]
+    """Model ID or Endpoint ID. Whitespace-only values are rejected."""
+
+    content: Required[List[ContentParam]]
+
+    dubbing_config: Optional[Optional[DubbingConfigParam]]
+    """Required for dubbing; forbidden in reference generation and separation."""
+
+    output_format: Optional[Optional[OutputFormat]]
+    """Output audio format. Omitted or null resolves to wav."""
+
+    audio_config: Optional[Optional[AudioConfigParam]]
+    """Optional output parameters. Defaults are applied by the service."""
+
+    include: Optional[Optional[List[Include]]]
+    """Additional artifacts; omitted/null/empty means none. video_url requires a video input. subtitles_source.* is only allowed for dubbing. Separation forbids a non-empty include."""
+
+    watermark: Optional[Optional[bool]]
+    """Append an audible rhythm watermark at the end. Defaults to false. Only supported in the cn-beijing deployment region; overseas deployments reject true with InvalidParameter (param=watermark)."""

src/arkruntime/types/audio_generation/audio_generation_response.py

@@ -0,0 +1,47 @@
+# Generated by datamodel-code-generator from ark-apis typespec/openapi.
+# DO NOT EDIT — regenerate with `make gen-py-<api>` in ark-apis.
+#
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from typing import Literal, Optional
+
+from arkruntime._models import BaseModel
+
+from .audio_config import AudioConfig
+from .effective_dubbing_config import EffectiveDubbingConfig
+from .output_format import OutputFormat
+from .response_content import ResponseContent
+from .usage import Usage
+
+
+class AudioGenerationResponse(BaseModel):
+    id: str
+    model: str
+    """
+    Model name/version. Reference generation, audio separation and video dubbing using an agent plan return the brand model identifier; if resolution fails, the request model is returned instead of the underlying routed model.
+    """
+    content: Optional[ResponseContent] = None
+    dubbing_config: Optional[EffectiveDubbingConfig] = None
+    """
+    Effective dubbing configuration, when applicable.
+    """
+    output_format: Optional[OutputFormat] = None
+    """
+    Dubbing returns the effective format; other scenes echo the supplied request field.
+    """
+    audio_config: Optional[AudioConfig] = None
+    """
+    Dubbing returns the effective sample rate; other scenes echo the supplied request config.
+    """
+    subtitle: Optional[Literal[False]] = None
+    """
+    Separation explicitly returns false. Other scenes currently omit this field; inspect content.subtitles for subtitle artifacts.
+    """
+    created: int
+    """
+    Creation time, Unix seconds.
+    """
+    usage: Optional[Usage] = None

src/arkruntime/types/audio_generation/audio_role.py

@@ -0,0 +1,14 @@
+# Generated by datamodel-code-generator from ark-apis typespec/openapi.
+# DO NOT EDIT — regenerate with `make gen-py-<api>` in ark-apis.
+#
+# Copyright (c) 2026 ByteDance Ltd. and/or its affiliates.
+# SPDX-License-Identifier: Apache-2.0
+
+from __future__ import annotations
+
+from enum import Enum
+
+
+class AudioRole(str, Enum):
+    reference_audio = "reference_audio"
+    separate_audio = "separate_audio"