localai-org-maint-bot c905d40d9e fix(reasoning): ignore preclosed prompt markers ...

Do not seed streaming reasoning state when the latest prompt thinking marker is already followed by its matching closing marker. This keeps direct Gemma 4 output in content when its template disables thinking with a preclosed channel.

Assisted-by: Codex:gpt-5

2026-07-29 14:02:20 +00:00

..

anthropic

feat: interleaved thinking with tool calls (reasoning_content alias + Anthropic thinking blocks) (#10744)

2026-07-08 16:45:43 +00:00

elevenlabs

feat(tts): support per-request instructions and params (#10172)

2026-06-04 11:45:02 +02:00

explorer

feat: add distributed mode (#9124)

2026-03-30 00:47:27 +02:00

jina

feat(whisper): honor client cancellation via ggml abort_callback (#9710)

2026-05-08 01:44:47 +02:00

localai

feat(classifier/VAD): support voice control on low power devices (#10804)

2026-07-29 12:50:22 +02:00

mcp

fix(mcp): bound MCP session connect so an unreachable server can't hang the widget (#10880) (#10884)

2026-07-22 15:49:40 +02:00

ollama

fix(ollama): set ContextSize via the embedded LLMConfig so the package builds (#11049)

2026-07-22 16:25:56 +02:00

openai

fix(reasoning): ignore preclosed prompt markers

2026-07-29 14:02:20 +00:00

openresponses

fix(reasoning): ignore preclosed prompt markers

2026-07-29 14:02:20 +00:00