CHANGELOG.md

Changelog

All notable changes to this project will be documented in this file.

[0.5.22]

Added

• IInputs now includes:
  • seedImage: Optional[Union[str, File]]

Changed

• imageInference now processes inputs.seedImage via _process_media before sending the request

[0.5.21]

Changed

• IVideoEditSettingsControls now uses flat fields for edit controls instead of nested per-control dataclasses:
  • poseStrength: Optional[str]
  • depthBlur: Optional[float]
  • normalsAugmentation: Optional[float]
  • trajectorySparsity: Optional[float]
  • face: Optional[bool]

Removed

• Per-control video edit dataclasses superseded by the flat IVideoEditSettingsControls fields:
  • IVideoEditPoseControl
  • IVideoEditDepthControl
  • IVideoEditNormalsControl
  • IVideoEditTrajectoryControl
  • IVideoEditFaceControl

[0.5.20]

Added

• ISettings now includes:
  • preserveInputSize: Optional[bool]
• Note in README: A new official SDK is available at runware-python. It covers the full inference and utility surface with REST or WebSocket transports and gets all new features going forward. This library remains supported during the transition.

[0.5.19]

Added

• Video editing/reframing helper dataclasses:
  • ISourcePosition
  • IVideoEditPoseControl
  • IVideoEditDepthControl
  • IVideoEditNormalsControl
  • IVideoEditTrajectoryControl
  • IVideoEditFaceControl
  • IVideoEditSettingsControls
  • IVideoEditSettings
• Text inference chat tools and message serialization:
  • ITextInferenceMessageTool (toolId, name, toolInput; serialized as id, name, input)
  • ITextInferenceMessage extends SerializableMixin with optional content, toolId, and tools; serialize() maps toolId → id on the wire

Changed

• ISettings now supports additional video parameters:
  • edit: Optional[Union[IVideoEditSettings, Dict[str, Any]]]
  • exrExport: Optional[bool]
  • hdr: Optional[bool]
  • loop: Optional[bool]
  • sourcePosition: Optional[Union[ISourcePosition, Dict[str, Any]]]
• Added dict coercion in ISettings.__post_init__ for:
  • edit → IVideoEditSettings
  • sourcePosition → ISourcePosition
• _buildTextRequest() now builds messages with ITextInferenceMessage.serialize() instead of dataclasses.asdict()
• ITextInferenceMessage.content is now Optional[str]
• instantiateDataclass now skips null values from API payloads instead of passing them through as explicit None
• instantiateDataclass fills missing required str fields with "" when the API leaves them out, so dataclasses like IModel can keep version: str and architecture: str as required fields
• Response dataclasses built via instantiateDataclass / instantiateDataclassList (IModel, IImage, IVideo, I3d, etc.) use a compact repr that omits unset, None, blank string, and empty additional_fields values

[0.5.18]

Added

• IDracoCompression dataclass for settings.dracoCompression:
  • enabled: Optional[bool]
  • level: Optional[int]
  • quantizationPosition: Optional[int]
  • quantizationNormal: Optional[int]
  • quantizationTexCoord: Optional[int]
• ISettings now includes:
  • dracoCompression: Optional[Union[bool, IDracoCompression, Dict[str, Any]]]

Changed

• IInputs.referenceImages now accepts Dict[str, Any] entries and coerces them to IInputReference in IInputs.__post_init__

[0.5.17]

Added

• IInputFont dataclass for inputs.fonts[]:
  • url: Optional[str]
  • text: Optional[str]
• IInputs now includes:
  • fonts: Optional[List[Union[IInputFont, Dict[str, Any]]]]
• IScoringRubric dataclass for settings.scoringRubric[]:
  • key: Optional[str]
  • label: Optional[str]
  • description: Optional[str]
  • weight: Optional[float]
  • passingScore: Optional[float]
  • scoreGuidance: Optional[List[Union[IScoringRubricScoreGuidance, Dict[str, Any]]]]
• IScoringRubricScoreGuidance (score, description)
• ISettings now includes:
  • enhancePrompt: Optional[bool]
  • backgroundMode: Optional[str]
  • scoringPrompt: Optional[str]
  • scoringRubric: Optional[List[Union[IScoringRubric, Dict[str, Any]]]]

Changed

• ISettings.promptEnhancement renamed to ISettings.promptEnhance

[0.5.16]

Added

• IPromptEnhancement dataclass for nested settings.promptEnhancement:
  • enabled: Optional[bool]
  • temperature: Optional[float]
  • topP: Optional[float]
• IInputFrame.timestamp: Optional[float] = None
• ISettings now includes:
  • sourceAudioSync: Optional[bool] = None
• Structured prompt support for imageInference requests via settings.structuredPrompt:
  • IStructuredPrompt (high_level_description, style_description, compositional_deconstruction)
  • IStructuredPromptCompositionalDeconstruction (background, elements)
  • IStructuredPromptElement (elementType, desc, text, bbox, color_palette); serializes elementType as API type
  • IStructuredPromptStyleDescription (aesthetics, lighting, photo, medium, art_style, color_palette)
  • ISettings.structuredPrompt: Optional[Union[IStructuredPrompt, Dict[str, Any]]] with dict coercion in ISettings.__post_init__
  • ISettings.copyrightDetection: Optional[bool]
• Image inference response fields on IImage:
  • status: Optional[str]
  • structuredPrompt: Optional[Dict[str, Any]] (returned as plain dict, matching API shape including type on elements)
• IImageInferenceOutputs.structuredPrompt: Optional[Dict[str, Any]] when the API nests it under outputs

Changed

• _requestVideo now preprocesses inputs.video the same way as inputs.frameImages
• Removed createImageFromResponse

[0.5.15]

Added

• IMeshCluster dataclass for settings.meshCluster:
  • thresholdConeHalfAngleRad: Optional[float]
  • refineIterations: Optional[int]
  • globalIterations: Optional[int]
  • smoothStrength: Optional[int]
• ISettings now includes:
  • textureFormat: Optional[str]
  • alphaMode: Optional[str]
  • remeshBand: Optional[float]
  • remeshProject: Optional[float]
  • meshCluster: Optional[Union[IMeshCluster, Dict[str, Any]]]

[0.5.14]

Added

• IUltralytics now includes:
  • inpaintSize: Optional[int]
• ISettings now includes:
  • creativity: Optional[str]
  • moodboards: Optional[List[Union[IMoodboard, Dict[str, Any]]]]
  • preserveAudio: Optional[bool]
• IMoodboard dataclass for settings.moodboards[] entries:
  • id: str
  • strength: Optional[float]
• IInputReference now includes:
  • role: Optional[str]

[0.5.13]

Added

• ISettings now includes:
  • autoCrop: Optional[bool]
  • chunkLength: Optional[int]
  • minChunkLength: Optional[int]
  • normalize: Optional[bool]
  • normalizeLoudness: Optional[bool]
  • latency: Optional[str]
  • conditionOnPreviousChunks: Optional[bool]
  • earlyStopThreshold: Optional[float]
• IAudioReferenceVoice (audio, text) for reference-voice input on audioInference
• IAudioInputs now includes:
  • referenceVoices: Optional[List[Union[IAudioReferenceVoice, Dict[str, Any]]]]
• IAudioSpeech.voices now accepts voice model ID strings in addition to IAudioVoice objects

Changed

• IAudioSpeech.volume type changed from Optional[int] to Optional[float]

[0.5.12]

Added

• IImageInference.outputType now supports Optional[Union[IOutputType, List[IOutputType]]].
• Image inference results now include outputs on IImage:
  • IImageInferenceOutputs with textBlocks: Optional[List[IImageInferenceTextBlock]]
  • IImageInferenceTextBlock (alignment, angle, color, fontAlternatives, fontName, fontSize, formatting, height, lineHeight, role, text, width, x, y)
• ISettings now includes:
  • fit: Optional[str]
  • caption: Optional[bool]
• IVideoSpeechSettings now includes:
  • volume: Optional[float]

[0.5.11]

Added

• Added support for taskType training.
• Added text inference cache support:
  • TextInferenceCacheScope = Literal["system", "system+history"]
  • TextInferenceCacheTtl = Literal["5m", "1h"]
  • ITextInferenceCache dataclass (scope, ttl) with request key cache
• ITextInputs now includes:
  • documents: Optional[List[Union[str, File]]]
• ITextInferenceTool now supports:
  • schema: Optional[Dict[str, Any]]
  • backward-compatible input_schema alias (serialized as schema when schema is not provided)
  • optional toolType (serialized as type)
• Added ITextInferenceToolChoice request serialization support:
  • request_key = "toolChoice"
  • toolType -> type mapping in serialize()
• ISettings now includes:
  • search: Optional[bool] = None
  • voicePrompt: Optional[str]
  • safetyFilter: Optional[bool]
  • cfgIntervalStart: Optional[float]
  • cfgIntervalEnd: Optional[float]
  • repaintMode: Optional[str]
  • repaintStrength: Optional[float]
• IAudioOutputFormat now supports:
  • ogg

Changed

• toolChoice moved from ISettings to the root ITextInference payload:
  • ITextInference.toolChoice: Optional[Union[ITextInferenceToolChoice, Dict[str, Any]]]
  • _buildTextRequest() now includes root-level toolChoice
• Unified text-input media preprocessing in runware/base.py via _processTextInputs() and applied it to both _requestText() and _requestTextStream(), so images/videos/documents are normalized consistently for sync/async/stream text requests.
• IVideoInference.speech now accepts either:
  • IVideoSpeechSettings
  • Dict[str, Any] (auto-coerced to IVideoSpeechSettings in __post_init__)

[0.5.10]

Added

• Added support for taskType getTaskDetails.
• Added IVideoReferenceImage dataclass with:
  • tag: Optional[str]
  • refType: Optional[str]
  • images: Optional[List[Union[str, File]]]
  • audio: Optional[Union[str, File]]
  • serialize() override that maps refType -> type in outgoing payloads.
• Added IVideoReferenceVideo dataclass with:
  • tag: Optional[str]
  • refType: Optional[str]
  • video: Optional[Union[str, File]]
  • serialize() override that maps refType -> type in outgoing payloads.
• Extended IVideoInputs to support structured references:
  • referenceImages: Optional[List[Union[str, File, IVideoReferenceImage]]]
  • referenceVideos: Optional[List[Union[str, IVideoReferenceVideo]]]
• Added process_media() in runware/utils.py as the canonical media normalization helper (supports local-file conversion and pass-through for URL/UUID/base64).
• Added backward-compat normalization in IVideoInputs.__post_init__:
  • dict inputs with type are accepted and internally mapped to refType for both referenceImages and referenceVideos.
• ISettings now includes Meshy 6 / 3D controls:
  • meshType: Optional[str]
  • topology: Optional[str]
  • decimation: Optional[int]
  • symmetry: Optional[str]
  • pose: Optional[str]
  • removeLighting: Optional[bool]
  • origin: Optional[str]
  • moderation: Optional[bool]
  • texturePrompt: Optional[str]
  • imageEnhancement: Optional[bool]
  • savePreRemeshedModel: Optional[bool]
• I3dInputs now includes:
  • referenceImages: Optional[List[Union[str, File]]]
• ISettings now includes:
  • keyframe: Optional[int]
• ISettings now includes additional 3dInference fields:
  • faceCount: Optional[int]
  • geometryOnly: Optional[bool]
  • generateType: Optional[str]
  • polygonType: Optional[str]
• I3dInference now includes:
  • negativePrompt: Optional[str]
• I3dOutputFormat now supports:
  • OBJ

Changed

• Improved SerializableMixin.serialize() to handle mixed list/tuple payloads by serializing SerializableMixin items item-by-item (instead of requiring all items to be mixin instances).
• Switched runware/base.py media normalization flow to canonical process_media() semantics (via _process_media delegation).
• Kept process_image() as a backward-compatible alias to process_media().
• Replaced type usage in video reference dataclasses with refType to avoid reserved-name ambiguity while preserving API payload compatibility (type is still sent on wire via serialize()).
• I3dOutputFormat now supports Meshy 6 target formats: GLB, OBJ, FBX, STL, USDZ, 3MF (while keeping PLY for compatibility).

[0.5.9]

Added

• ISettings now includes:
  • syncMode: Optional[str]
  • mode: Optional[str]
  • emotion: Optional[str]
  • activeSpeakerDetection: Optional[Union[IActiveSpeakerDetection, Dict[str, Any]]]
  • occlusionDetection: Optional[bool]
  • segments: Optional[List[Union[ISegment, Dict[str, Any]]]]
  • tts: Optional[Union[ITTSSettings, Dict[str, Any]]]
  • colorCorrection: Optional[bool]
  • renderingSpeed: Optional[str]
  • magicPrompt: Optional[str]
  • background: Optional[str]
• Added IActiveSpeakerDetection (autoDetect, frameNumber, coordinates, boundingBoxes) with validation for boundingBoxes entries (must be null or four numeric values).
• Added ISegment for Sync segments (startTime, endTime, audio, ref, audioStartTime, audioEndTime).
• Added ITTSSettings for Sync TTS tuning (stability, similarityBoost).
• Added compatibility aliases:
  • ISyncSegment = ISegment
  • ISyncActiveSpeakerDetection = IActiveSpeakerDetection
  • ISyncSettings = ISyncProviderSettings
• IAudioSpeech now includes:
  • voices: Optional[List[Union[IAudioVoice, Dict[str, Any]]]]
• Added IAudioVoice for multi-speaker TTS mapping:
  • speaker: str
  • voice: str
• IImageInference now includes:
  • upscaleFactor: Optional[float]

Changed

• Updated ISyncProviderSettings to include Sync 3 fields:
  • mode: Optional[str]
  • emotion: Optional[str]
  • activeSpeakerDetection: Optional[Union[IActiveSpeakerDetection, Dict[str, Any], bool]]
  • occlusionDetection: Optional[bool]
  • segments: Optional[List[Union[ISegment, Dict[str, Any]]]]
• Added post-init normalization for Sync settings so dict inputs are coerced into IActiveSpeakerDetection, ISegment, and ITTSSettings objects.

[0.5.8]

Added

• ISettings now includes:
  • useOriginalAlpha: Optional[bool]
  • material: Optional[str]
  • polyCount: Optional[float]
  • taPose: Optional[bool]
  • boundingBox: Optional[List[int]]
  • meshMode: Optional[str]
  • addons: Optional[List[str]]
  • hdTexture: Optional[bool]
  • instrumental: Optional[bool]
  • lyricsOptimizer: Optional[bool]
• IAudioInference now includes:
  • outputQuality: Optional[int]
• ITextInputs now includes:
  • videos: Optional[List[Union[str, File]]]
• IVideoInputs now includes:
  • referenceAudios: Optional[List[str]]

Changed

• Fixed httpx version dependency during installation in setup.py.
• setup.py: find_packages excludes the top-level tests package and its subpackages.

[0.5.7]

Added

• Streaming support for LLMs.
• ISettings now includes:
  • includePrefix: Optional[bool]
  • audioTemperature: Optional[float]
  • topK: Optional[int]
• IAudioInputs now includes:
  • audios: Optional[List[str]]

[0.5.6]

Added

• ISettings now includes:
  • imageAutoFix: Optional[bool]
  • faceLimit: Optional[int]
  • texture: Optional[bool]
  • pbr: Optional[bool]
  • textureSeed: Optional[int]
  • textureAlignment: Optional[str]
  • textureQuality: Optional[str]
  • autoSize: Optional[bool]
  • orientation: Optional[str]
  • quad: Optional[bool]
  • compress: Optional[str]
  • smartLowPoly: Optional[bool]
  • generateParts: Optional[bool]
  • exportUv: Optional[bool]
  • geometryQuality: Optional[str]
  • editRegions: Optional[List[List[List[int]]]]
  • sequential: Optional[bool]
  • thinking: Optional[bool]
  • colorPalette: Optional[List[Union[IColorPaletteEntry, Dict[str, Any]]]]
  • style: Optional[str]
  • thinking: Optional[str]
  • multiClip: Optional[bool]
  • voiceDescription: Optional[str]
  • Image upscale: enhanceDetails, realism, and existing upscale tuning fields on ISettings (steps, seed, CFGScale, prompts, scheduler, etc.)
• I3dInputs now includes images: Optional[List[Union[str, File]]].
• IVideoInputs.audios for video inference.
• IImageUpscale: targetMegapixels, optional upscaleFactor, deliveryMethod (default "sync"); IImageUpscale.settings typed as ISettings.
• IUpscaleSettings: deprecated subclass of ISettings (deprecation warning on use).
• _upscaleGan: async deliveryMethod handling for IAsyncTaskResponse + polling.
• New dataclass IColorPaletteEntry: hex: str, ratio: Optional[Union[str, float]].
• IInputs now includes images: Optional[List[Union[str, File]]].

Changed

• _pollResults: imageUpscale, vectorize, and imageBackgroundRemoval resolve to IImage (fixes getResponse returning IVideo without image URLs).
• _upscaleGan: conditional upscaleFactor / targetMegapixels; forward deliveryMethod.

[0.5.5]

Changed

• Relax aiofiles and python-dotenv pins to >= minimums in setup.py and requirements.txt so the package installs cleanly alongside environments that pin newer compatible releases.

[0.5.4]

Added

• Added runware/version.py as the canonical __version__ string; runware/__init__.py imports it so from runware import __version__ matches the package version.
• WebSocket connections send additional headers on connect: X-SDK-Name: python and X-SDK-Version: <__version__> (see runware/server.py).

[0.5.3]

Added

• Added width: Optional[int], height: Optional[int], and fps: Optional[int] to IVideoUpscale.
• Added IElements dataclass for video inference inputs.elements[] with:
  • id: Optional[str]
  • description: Optional[str]
  • frontalImage: Optional[Union[str, File]]
  • images: Optional[List[Union[str, File]]]
  • videos: Optional[List[str]]
  • voice: Optional[List[str]]
  • tags: Optional[List[str]]
• Added elements: Optional[List[Union[IElements, Dict[str, Any]]]] to IVideoInputs.
• IAudioOutputFormat extended with wav, mp3, pcm, opus, aac, flac (existing MP3 retained).
• ISettings: maxNewTokens, transcript, xVectorOnly for reference-audio / ICL TTS flows.
• IAudioInputs: audio (reference audio URL or data URI).
• IAudioSpeech: language (e.g. input language for TTS).

Changed

• Updated _requestVideoUpscale() to forward width, height, and fps when provided.
• Updated IVideoInputs.__post_init__ to coerce inputs.elements dictionary items into IElements instances.

[0.5.2]

Added

• Added photoMaker as a nested object in imageInference using IPhotoMakerSettings; IImageInference.photoMaker: Optional[Union[IPhotoMakerSettings, Dict[str, Any]]]. Nested settings use images/inputImages + style/strength. Standalone taskType="photoMaker" remains a separate operation via IPhotoMaker.
• Added IIpAdapter.guideImages, combineMethod, weightType, embedScaling, weightComposition; guideImage optional; base supports guideImages with process_image.
• Added IEmbedding.weight: Optional[float].
• Added dict coercion in IImageInference.post_init for photoMaker, instantID, acePlusPlus, puLID, ultralytics, outpaint, refiner; list coercion for embeddings (dict -> IEmbedding) and ipAdapters (dict -> IIpAdapter).
• Added IInputReference.type and IInputReference.strength (for sketch-style refs).
• Added image inference support so inputs.referenceImages can be IInputReference items; each image is run through process_image before send.
• Added IGoogleProviderSettings.safetyTolerance.
• Added bpm, keyScale, timeSignature, vocalLanguage, coverConditioningScale, repaintingStart, and repaintingEnd to ISettings.
• Added audio: Optional[str] to IAudioInputs.
• Added IGoogleProviderSettings.resizeMode.

Changed

• IImageInference: refiner, outpaint, instantID, acePlusPlus, puLID, ultralytics, photoMaker typed as Union[Type, Dict[str, Any]]; embeddings as List[Union[IEmbedding, Dict[str, Any]]]; ipAdapters as List[Union[IIpAdapter, Dict[str, Any]]].

[0.5.1]

Added

• IVectorize: width, height, positivePrompt, providerSettings (optional).
• IVectorize.inputs optional (prompt-only vectorize e.g. recraft:v4@vector).
• VectorizeProviderSettings type alias (IRecraftProviderSettings) in types.
• _processVectorizeInputs and _buildVectorizeRequest in base (vectorize follows video/3D flow).
• Single _addProviderSettings(request_object, payload) for image, image background removal, image upscale, vectorize, video, audio, text (replaces _addImageProviderSettings, _addAudioProviderSettings, _addTextProviderSettings).
• Added post_init to IAudioInference dataclass for speech = Optional[Union[IAudioSpeech, Dict[str, Any]]].
• Added textNormalization: Optional[bool] = None to ISettings dataclass.
• Image IAdvancedFeatures: layerDiffuse, hiresFix, watermark (IWatermark), and regionalPrompting (IRegionalPrompting with regions: List[IRegion]).
• Video IVideoAdvancedFeatures: IWatermark (text/image, position, opacity, colors) and watermark field.
• Serializable advanced features: IFluxKontext, IRegion, IRegionalPrompting, IWatermark, IAdvancedFeatures, and IVideoAdvancedFeatures use SerializableMixin for request serialization.
• Added avatar: Optional[str], background: Optional[str] to IVideoInputs.
• Added speed: Optional[float], pitch: Optional[float], language: Optional[str] to IVideoSpeechSettings.
• Added expressiveness: Optional[str], removeBackground: Optional[bool], backgroundColor: Optional[str] to ISettings.

Changed

• Vectorize: inputs resolved via process_image when present; optional and added with _addOptionalField.
• _addProviderSettings payload type: Union[IImageInference, IImageBackgroundRemoval, IImageUpscale, IVectorize, IVideoInference, IAudioInference, ITextInference].
• Request wiring: Image/video/audio builders now rely on _addOptionalBuiltInDataTypesFields() instead of hardcoded optional field lists, ensuring fields like acceleration and webhookURL are always forwarded when set.

[0.5.0]

Changed

• Replace polling with asyncio.Future for O(1) message routing via _pending_operations HashMap
• Add semaphore to limit concurrent requests (default 15, env RUNWARE_MAX_CONCURRENT_REQUESTS)
• Add jitter on reconnect to prevent thundering herd
• Increase PING_TIMEOUT_DURATION from 10s to 30s for stability
• Graceful cancellation of pending operations on disconnect

Added

• Added post_init to IImageInference, IImageCaption, IImageUpscale, IVideoInference, I3dInference, IAudioInference, IVideoBackgroundRemoval
• Added promptExtend: Optional[bool] to ISettings
• Added IBytedanceProviderSettings.optimizePromptMode: Optional[str]
• Added settings to IVideoInference dataclass
• Added draft: Optional[bool], audio: Optional[bool], and promptUpsampling: Optional[bool] to ISettings dataclass
• Added post_init to ISettings dataclass

[0.4.47]

Added

• Added IRGB dataclass (rgb: List[int]) with validation (exactly 3 elements, each 0–255)
• Added IRecraftProviderSettings with styleId: Optional[str], colors: Optional[List[IRGB]], backgroundColor: Optional[IRGB]; added as ImageProviderSettings.
• Added ISettings.lyrics: Optional[str] and ISettings.guidanceType: Optional[str].
• Added negativePrompt, seed, steps, CFGScale, and settings to IAudioInference dataclass.
• Added IAudioSpeech dataclass (text, voice, speed, volume, pitch, emotion, tone: Optional[List[str]]).
• Added channels: Optional[int] to IAudioSettings.
• Added speech: Optional[IAudioSpeech] and settings: Optional[ISettings] to IAudioInference.
• Added languageBoost: Optional[str] and turbo: Optional[bool] to ISettings.
• Added edit, color, lightDirection, lightType, season (all Optional[str]) to IBriaProviderSettings.

[0.4.46]

Added

• Added IKlingMultiPrompt: new dataclass with prompt: str and duration: float for Kling multiPrompt payloads.

Changed

• fileToBase64(): allow application/octet-stream as MIME type for .glb and .ply files when MIME cannot be guessed.
• Recursive serialization in SerializableMixin.serialize(): replaced flat asdict + None filter with a recursive implementation that:
  • Skips None and keys starting with _
  • Recursively serializes nested SerializableMixin instances
  • Serializes lists/tuples of SerializableMixin as lists of dicts (e.g. for multiPrompt and other list payloads).

[0.4.45]

Added

• Added ISparseStructure, IShapeSlat, ITexSlat (guidanceStrength, guidanceRescale, steps, rescaleT).
• Added I3dInference.settings: Optional[ISettings] and outputQuality: Optional[int].
• Added I3dInputs.meshFile: Optional[Union[str, File]].

Changed

• Extended ISettings with 3D fields: textureSize, decimationTarget, remesh, resolution, sparseStructure, shapeSlat, texSlat.

[0.4.44]

Added

• Added support for textInference task type (TEXT_INFERENCE in ETaskType enum)
• Added ITextInferenceMessage with role and content for chat messages
• Added ITextInferenceUsage for token usage (promptTokens, completionTokens, totalTokens, thinkingTokens)
• Added IGoogleTextProviderSettings with thinkingLevel for Gemini
• Added TextProviderSettings type alias for text inference provider settings
• Added ITextInference for requests (model, messages, taskUUID, deliveryMethod, maxTokens, temperature, topP, topK, seed, stopSequences, includeCost, providerSettings)
• Added IText for responses (taskType, taskUUID, text, finishReason, usage, cost, status)
• Added textInference() to Runware
• Added _buildTextRequest, _requestText, _handleInitialTextResponse in base
• Added _addTextProviderSettings for provider settings (matching image/video/audio pattern)
• Added getResponse() support for List[IText] when polling textInference tasks
• Added TEXT_INITIAL_TIMEOUT for async delivery (configurable via RUNWARE_TEXT_INITIAL_TIMEOUT)
• Added TEXT_POLLING_DELAY for polling cadence (configurable via RUNWARE_TEXT_POLLING_DELAY)
• Added IVideoInference.scheduler: Optional[str] = None

Changed

• Enabled async delivery for text inference (returns IAsyncTaskResponse and uses getResponse() for polling)

[0.4.43]

Added

• Added support for "3dInference" task type
• Added I3dInputs dataclass with image and mask parameters for 3D inference inputs
• Added I3dOutputFormat type: Literal["GLB", "PLY"]
• Added I3dInference dataclass for 3D inference requests (model, positivePrompt, seed, inputs, outputFormat, etc.)
• Added IObject dataclass with uuid and url for 3D output files
• Added I3dOutput dataclass with files: Optional[List[IObject]] = None for 3D output files in responses
• Added I3d dataclass for 3D inference responses (taskType, taskUUID, cost, status, seed, outputs)
• Added inference3d() method to Runware for 3D inference
• Added getResponse() support for List[I3d] when polling 3dInference tasks
• Added _handleInitial3dResponse, _process3dInputs, _build3dRequest, _request3d in base
• Added IUltralytics dataclass with the following parameters:
  • maskBlur: Optional[int]
  • maskPadding: Optional[int]
  • confidence: Optional[float]
  • positivePrompt: Optional[str]
  • negativePrompt: Optional[str]
  • steps: Optional[int]
  • CFGScale: Optional[float]
  • strength: Optional[float]
• Added IImageInference.ultralytics: Optional[IUltralytics]
• Added quality: Optional[str] = None to ISettings
• Added audio: Optional[bool] = None to IViduProviderSettings

Changed

• Split IOutput into media-specific output types: IOutput (video: draftId, videoId) and I3dOutput (3D: files)

[0.4.42]

Added

• Added IOutput dataclass with the following parameters:
  • draftId: Optional[str] = None
  • videoId: Optional[str] = None
• Added IVideo.outputs: Optional[IOutput] = None to support draft ID in video responses
• Added IVideoInputs.draftId: Optional[str] = None for passing draft task ID in video inference inputs for bytedance:seedance@1.5-pro
• Added IVideoInputs.videoId: Optional[str] = None for passing video ID in video inference inputs for openai:3@1
• Added IBytedanceProviderSettings.draft: Optional[bool] = None to enable draft mode for Seedance video generation
• Added IInputs.mask: Optional[Union[str, File]] = None
• Added ISourcefulProviderSettings with the following parameters:
  • transparency: Optional[bool] = None
  • enhancePrompt: Optional[bool] = None
  • fontInputs: Optional[List[Dict[str, Any]]] = None
• Added IInputs.superResolutionReferences: Optional[List[Union[str, File]]] = None for super resolution guidance in image-to-image inference

Changed

• Enhanced instantiateDataclass() function to automatically handle nested dataclasses

[0.4.41]

Changed

• Updated IBriaProviderSettings.preserveAudio default from Optional[bool] = True to None
• Updated IBriaProviderSettings.autoTrim default from Optional[bool] = False to None

Added

• Added taskUUID attribute to IUploadModelBaseType to allow passing taskUUID from user in modelUpload() method

[0.4.40]

Added

• Added _retry_with_reconnect() method that wraps all public API methods for automatic reconnection on authentication errors (up to 10 retries)
• Added MAX_RETRY_ATTEMPTS constant in runware/utils.py for retry mechanism configuration
• Added IMireloProviderSettings with startOffset: Optional[int] = None parameter
• Added ISettings.layers: Optional[int] = None parameter
• Added ISettings.trueCFGScale: Optional[float] = None parameter
• Added IAudio.videoUUID: Optional[str] = None field
• Added IAudio.videoURL: Optional[str] = None field
• Added IAudio.seed: Optional[int] = None field

Fixed

• Fixed duplicate heartbeat tasks by cancelling existing task before creating new one in runware/server.py
• Fixed connection loss detection during request waits with session UUID tracking
• Fixed conflictTaskUUID error handling to extract taskType and deliveryMethod from error context instead of relying on global message state

Changed

• Refactored retry logic: Removed redundant asyncRetry calls from 9 methods and them wrapped with _retry_with_reconnect

[0.4.39]

Added

• Added MODEL_UPLOAD_TIMEOUT constant in runware/utils.py with 15-minute default (900000ms) for large model uploads
• Added RUNWARE_MODEL_UPLOAD_TIMEOUT environment variable support for configuring model upload timeout

Changed

• Updated _modelUpload() in runware/base.py to use MODEL_UPLOAD_TIMEOUT instead of general timeout to prevent timeouts during large LoRA model uploads

[0.4.38]

Added

• Added IWanAnimate dataclass with the following parameters:
  • mode: Optional[str] = None
  • retargetPose: Optional[bool] = None
  • prevSegCondFrames: Optional[int] = None
• Added VideoAdvancedFeatureTypes union type: VideoAdvancedFeatureTypes = IWanAnimate
• Updated IVideoAdvancedFeatures to support advancedFeature: Optional[VideoAdvancedFeatureTypes] = None for incorporating WanAnimate features
• Added IVideoInputs.referenceVoices: Optional[List[str]] = None
• Added IKlingAIProviderSettings.characterOrientation: Optional[str] = None

Fixed

• Fixed NoneType error when initial requests timeout by returning IAsyncTaskResponse object instead of processing None values
• Added error handling in _handleInitialVideoResponse, _handleInitialImageResponse, and _handleInitialAudioResponse to handle empty/None initial responses
• Added validation in instantiateDataclassList() method to raise descriptive error when data_list is None or empty
• Improved error messages for timeout scenarios with detailed context including TaskUUID and delivery method

Changed

• Refactored serialization: Removed BaseRequestField class and moved to_request_dict() method to SerializableMixin
• Updated multiple dataclasses to inherit from SerializableMixin instead of BaseRequestField:
  • IAcceleratorOptions
  • ISafety
  • ISettings
  • IInputs
  • IVideoInputs
  • IAudioSettings
  • IVideoSpeechSettings
  • IAudioInputs
• Updated IVideoAdvancedFeatures.serialize() method to properly handle advancedFeature serialization using to_request_dict()

[0.4.37]

Added

• Added audio parameter to IBytedanceProviderSettings (providerSettings.bytedance.audio)
• Added new parameters to IBriaProviderSettings:
  • preserveAudio
  • autoTrim
• Added IBriaMaskSettings with the following parameters:
  • foreground
  • prompt
  • frameIndex
  • keyPoints

Changed

• Added validation in _pollResults(): if number_results is None, then number_results = 1

[0.4.36]

Added

• Added IBlackForestLabsProviderSettings with safetyTolerance parameter
• Added new ISettings class with the following parameters:
  • temperature
  • systemPrompt
  • topP
• Added search parameter to IGoogleProviderSettings
• Added new parameters to IAlibabaProviderSettings:
  • promptExtend
  • audio
  • shotType

Changed

• Updated IInputs.references default value from field(default_factory=list) to None
• Updated IVideoInputs.references default value from field(default_factory=list) to None

[0.4.35]

Added

• Added provider setting for Alibaba: alibaba.promptEnhancer
• Added IAudioInput and ISpeechInput to IVideoInputs for video inference
• Added ISyncProviderSettings to video inference
• Added sound parameter to IKlingAIProviderSettings (providerSetting.Kling.sound)

Changed

• Updated referenceImages type from Optional[List[IInputReference]] to Optional[List[Union[str, File, IInputReference]]] in IInputs
• Renamed IPixverseSpeechSettings to IVideoSpeechSettings with backward compatibility
• Made numberResults optional in videoInference