Kirill Turanskiy 2ea95266e3 fix: clamp reasoning_effort to valid OpenAI-format values ...

CPA-internal thinking levels like 'xhigh' and 'minimal' are not accepted
by OpenAI-format providers (MiniMax, etc.). The OpenAI applier now maps
non-standard levels to the nearest valid reasoning_effort value before
writing to the request body:

  xhigh   → high
  minimal → low
  auto    → medium

2026-02-18 03:36:42 +03:00

..

antigravity

fix(thinking): ensure includeThoughts is false for ModeNone in budget processing

2026-02-05 10:24:42 +08:00

claude

feat(claude): clamp thinking budget to max_tokens constraints

2026-01-19 16:32:20 +08:00

codex

fix(thinking): improve model lookup and validation

2026-01-15 13:06:40 +08:00

gemini

fix(thinking): ensure includeThoughts is false for ModeNone in budget processing

2026-02-05 10:24:42 +08:00

geminicli

fix(thinking): ensure includeThoughts is false for ModeNone in budget processing

2026-02-05 10:24:42 +08:00

iflow

feat(thinking): enable thinking toggle for qwen3 and deepseek models

2026-01-28 09:54:05 +08:00

kimi

Add Kimi (Moonshot AI) provider support

2026-02-05 19:24:46 -05:00

openai

fix: clamp reasoning_effort to valid OpenAI-format values

2026-02-18 03:36:42 +03:00