82030615 da0046550a fix(responses): report usage on response.completed so clients can auto-compact ...

Map upstream Chat Completions usage to the Responses API shape and attach it to response.completed. Capture chunk.usage before the empty-choices guard so the usage-only trailer chunk survives, and defer completion to flushEvents() when usage is not yet known — only on the direct openai:openai-responses route, since a pivoted stream never reaches flushEvents. Fixes #3432.

2026-09-21 21:27:01 +07:00

..

concerns

fix(translator): drop replayed reasoning fields for Groq/Mistral/Cerebras (#4220)

2026-09-21 20:02:21 +07:00

formats

fix(translator): keep tool-result images, restore Kiro tool names, preserve thinking display

2026-09-17 18:12:05 +07:00

request

fix(antigravity): scope cached thought signatures to the model family

2026-09-18 17:09:36 +07:00

response

fix(responses): report usage on response.completed so clients can auto-compact

2026-09-21 21:27:01 +07:00

schema

fix(claude): drop server_tool_use blocks carrying a foreign id

2026-09-02 20:07:16 +07:00

formats.js

feat: add CommandCode provider support

2026-05-07 23:01:33 +07:00

index.js

fix(opencode): cloak Responses requests that already have tools

2026-09-19 10:15:45 +07:00