refactor: centralize CoT parsing in backend for streaming mode#16394

Merged

ggerganov merged 11 commits intoggml-org:masterfrom

ServeurpersoCom:streaming-aware-cpp-parser

Oct 8, 2025

Commits on Oct 6, 2025

refactor: unify reasoning handling via backend reasoning_content, drop frontend tag parsing
ServeurpersoCom
committed
refactor: implement streaming-aware universal reasoning parser
ServeurpersoCom
committed
refactor: address review feedback from allozaur

ServeurpersoCom
and
allozaur
committed
refactor: restore forced reasoning prefix to pass test-chat ([chat] All tests passed)
ServeurpersoCom
committed
refactor: address review feedback from ngxson
ServeurpersoCom
committed
debug: say goodbye to curl -N, hello one-click raw stream
ServeurpersoCom
committed
Update tools/server/webui/src/lib/components/app/chat/ChatMessages/ChatMessage.svelte

ServeurpersoCom
and
allozaur
committed
webui: add Storybook example for raw LLM output and scope reasoning format toggle per story
ServeurpersoCom
committed
npm run format
ServeurpersoCom
committed
chat-parser: address review feedback from ngxson

ServeurpersoCom
and
ngxson
committed

Commits on Oct 7, 2025

Merge branch 'master' into streaming-aware-cpp-parser
allozaur
authored