Release v0.10.2

Release v0.10.2

Changes

• feat(huggingface): add text generation provider (#183)

Installation

uv add "celeste-ai==0.10.2"

Release v0.10.1

Release v0.10.1

Changes

• feat(models): add Seedream 5.0 Lite and fix BytePlus presets (#181)

Installation

uv add "celeste-ai==0.10.1"

Release v0.10.0

Release v0.10.0

Changes

• fix(openai): route streaming image edits to correct endpoint and format (#180)
• feat(models): add Gemini 3.1 Pro and Claude Sonnet 4.6 (#179)
• release: v0.10.0 (#177)
• refactor: consolidate media constraints via generic base classes (#175) (#176)
• fix: align _parse_usage return types with RawUsage contract (#174)
• refactor: consolidate init, _build_stream_metadata, _parse_output into base Stream (#169) (#172)
• fix: catch UnicodeDecodeError in _handle_error_response (#170) (#171)
• refactor: deduplicate streaming methods across all modalities (#168)
• refactor: remove _build_headers() side effect in Anthropic client (#159) (#163)
• refactor: add _json_headers() helper to APIMixin (#162)
• refactor: add FieldMapper base class and migrate 41 simple mappers (#161)
• refactor: extract _content_fields ClassVar to eliminate duplicate _build_metadata() overrides (#160)
• refactor: extract Chat Completions protocol from 4 providers (#154)
• refactor: migrate OpenResponses to protocol layer (#153)
• fix: use strict=False in json.loads for structured output parsing (#152)
• refactor: centralize usage/finish_reason type conversion in base classes (#151)
• feat: add auth passthrough to all namespace methods (#146)
• feat: add base64 serialization config for Output and Chunk models (#145)
• ci: fix Opus 4.6 model config for claude-code-action v1 (#141)
• feat: add web search support for Anthropic models (#140)
• refactor: let Artifact validator handle base64 decoding across all providers (#139)
• feat: add Vertex AI support for all providers (#135)
• Add Claude Code GitHub Workflow (#136)
• chore(deps): make google-auth optional under celeste-ai[gcp] (#133)
• chore(deps): remove requests and ipykernel from runtime deps (#132)
• fix: rewrite CONTRIBUTING.md with accurate workflow (#131)

Installation

uv add "celeste-ai==0.10.0"

Release v0.9.7

Release v0.9.7

Changes

• feat(anthropic): add Claude Opus 4.6 model (#130)

Installation

uv add "celeste-ai==0.9.7"

Release v0.9.6

Release v0.9.6

Changes

• fix(xai): use ImageUrl struct format for image edits (#129)
• feat(xai): add Grok Imagine for image and video generation (#128)

Installation

uv add "celeste-ai==0.9.6"

v0.9.5

feat: expose extra_body parameter on all modalities (#126)

* feat: expose extra_body parameter on all modalities

Add extra_body parameter to all public methods across images, audio,
videos, and embeddings modalities. This allows users to pass
provider-specific request fields (e.g., Google's generationConfig,
imageConfig) without resorting to private methods.

Updated methods:
- images: generate, edit (stream, sync, sync.stream)
- audio: speak (stream, sync, sync.stream)
- videos: generate (sync)
- embeddings: embed (async, sync)

Also updated the modality client template for future modalities.

Fixes #124

https://claude.ai/code/session_01KYduqFZTvWMNMBW9b1nLXF

* style: format with ruff

https://claude.ai/code/session_01KYduqFZTvWMNMBW9b1nLXF

---------

Co-authored-by: Claude <noreply@anthropic.com>

v0.9.4

Fixed

• Fix gemini-3-pro-image-preview failing with 400 error in celeste.images.edit and celeste.images.generate by using correct responseModalities (#123)

What's Changed

• Fix Gemini response modalities to include TEXT and IMAGE by @Kamilbenkirane in #125

Full Changelog: v0.9.3...v0.9.4

Release v0.9.3

Release v0.9.3

Changes

• feat(images): add Ollama images provider with flexible parameter validation (#119)
• Rename title field in MovieReview class (#117)

Installation

uv add "celeste-ai==0.9.3"

Release v0.9.2

Release v0.9.2

Changes

• fix: strip whitespace from API keys to prevent HTTP header errors (#116)
• docs: add Colab quickstart notebook (#115)
• Update provider logos to match implemented providers (#113)
• update readme for v1 public beta launch (#112)

Installation

uv add "celeste-ai==0.9.2"

Release v0.9.1

Release v0.9.1

Changes

• feat: Add multi-turn conversation support and OpenResponses/Ollama providers (#111)

Installation

uv add "celeste-ai==0.9.1"