masterlyj b94bc18e59 fix: treat reasoning_effort="none" as thinking disabled and route gemma to Gemini provider ...

- Do not send reasoning_effort="none" to APIs (prevents 400 on gemma/Gemini)
- Treat "none" as thinking disabled in thinking_style, Kimi, and reasoning_content backfill paths
- Fix Anthropic extended thinking not respecting "none"
- Fix Azure OpenAI temperature suppression and reasoning body for "none"
- Fix Codex reasoning body for "none"
- Add "gemma" keyword to Gemini ProviderSpec for correct auto routing

2026-04-29 15:41:11 +08:00

..

openai_responses

refactor(providers): rename openai responses helpers

2026-04-02 13:43:34 +08:00

__init__.py

fix: restore GitHub Copilot auth flow

2026-04-02 03:46:40 +08:00

anthropic_provider.py

fix: treat reasoning_effort="none" as thinking disabled and route gemma to Gemini provider

2026-04-29 15:41:11 +08:00

azure_openai_provider.py

fix: treat reasoning_effort="none" as thinking disabled and route gemma to Gemini provider

2026-04-29 15:41:11 +08:00

base.py

refactor(codex): name progress delta capability semantically

2026-04-27 18:48:05 +08:00

factory.py

feat(providers): add extra_body config for OpenAI-compatible endpoints

2026-04-28 15:56:13 +08:00

github_copilot_provider.py

fix: refresh copilot token before requests

2026-04-02 03:46:40 +08:00

openai_codex_provider.py

fix: treat reasoning_effort="none" as thinking disabled and route gemma to Gemini provider

2026-04-29 15:41:11 +08:00

openai_compat_provider.py

fix: treat reasoning_effort="none" as thinking disabled and route gemma to Gemini provider

2026-04-29 15:41:11 +08:00

registry.py

fix: treat reasoning_effort="none" as thinking disabled and route gemma to Gemini provider

2026-04-29 15:41:11 +08:00

transcription.py

fix(transcription): harden language parameter validation and tests

2026-04-22 12:41:32 +08:00