Benjamin Admin 0106f3b5b6 fix: use Ollama directly for correction generation (bypass SDK think-mode) ...

SDK LLM chat returns empty content due to Qwen think-mode. Direct Ollama
/api/generate call with stream:false gets the full response including
think tags which we strip.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-04-28 16:30:51 +02:00

11 KiB

Raw Blame History

View Raw