v0.1.43

Changelog

Bug Fixes

• 9025183 fix(streaming): normalize non-SSE upstream chat completions (#412)

Performance

• 2677c1f perf: faster model resolution, JSON decoding, and request snapshotting (#413)

v0.1.42

Changelog

Features

• faf65b1 feat(dashboard): support fully offline use (#408)

v0.1.41

Changelog

Features

• 9115c3b feat(cli): add --health probe and Docker HEALTHCHECK (#404)
• 85c772b feat(server): add /health/ready readiness endpoint (#405)

Bug Fixes

• d24d71b fix(embeddings): honor encoding_format + fail loudly on empty ollama embeddings (#406)

v0.1.40

Changelog

Features

• 48fee93 feat(bailian): add native support for Alibaba Cloud Bailian (百炼) (#392)
• 284043d feat(opencode_go): add native OpenCode Go (OpenCode Zen) provider (#395)
• 72bf0a1 feat(realtime): add Azure OpenAI and Z.ai/Zhipu GLM realtime providers (#396)
• aa9cc0f feat(realtime): add OpenAI Realtime voice via websocket reverse proxy (#394)

Bug Fixes

• 70e5301 fix(bailian): preserve passthrough bodies (#393)

v0.1.39

Changelog

Features

• 414b491 feat(xiaomi): add Xiaomi MiMo provider with audio endpoint translation (#389)

Bug Fixes

• fb7e1cd fix(cache): route go-redis internal logs through slog (#391)

v0.1.38

Changelog

Bug Fixes

• e6aff03 fix(anthropic): map claude-fable-5 reasoning to adaptive thinking (#385)
• 8b79b38 fix(usage): cost text-to-speech by output audio duration (#376)
• e4373a9 fix: resolve gateway issues found by end-to-end testing (#386)

v0.1.37

Changelog

Features

• b31d3f8 feat(agents): support OpenAI Agents SDK (#353)
• ee3ee6d feat(usage): track and cost STT/TTS interactions (#372)

Bug Fixes

• efe4ad8 fix: extract system messages from messages array in Anthropic API (#374)

v0.1.36

Changelog

Features

• b22afc4 feat(audit): capture and play back transcription input audio (#365)
• d8f5eff feat(audit): log audio endpoint bodies with dashboard playback (#364)

v0.1.35

Changelog

Features

• 60dc00c feat(anthropic): support Claude Opus 4.8 effort levels and adaptive thinking (#360)
• 06a3869 feat(audio): add OpenAI-compatible text-to-speech and transcription endpoints (#362)

Bug Fixes

• 670d47f fix(release): publish draft release without checkout (#359)

v0.1.34

Changelog

Features

• 581cf4a feat(server): add OpenAI Conversations API endpoints (#347)

Bug Fixes

• 666066b fix(providers): refresh unavailable providers on request (#357)