.env.example

API_BASE=http://gx10.aquantico.lan:8093
OLLAMA_BASE_URL=http://gx10.aquantico.lan:11434
OLLAMA_MODEL=qwen3.5:9b
OLLAMA_NUM_PREDICT=4096
chore(diarization-ui): default API_BASE to gx10 and use fully qualified nginx image for podman 2026-03-21 13:24:11 +01:00			`API_BASE=http://gx10.aquantico.lan:8093`
feat(diarization-ui): add FastAPI UI backend with sqlite storage and Ollama analysis pipeline 2026-03-21 13:47:36 +01:00			`OLLAMA_BASE_URL=http://gx10.aquantico.lan:11434`
			`OLLAMA_MODEL=qwen3.5:9b`
fix(diarization-ui): prevent repetition loops in Ollama generation Adds repeat_penalty=1.15 and repeat_last_n=128 to suppress token repetition loops (e.g. "tragen" -> "tragen" -> ...). Also caps output via num_predict (default 4096, configurable via OLLAMA_NUM_PREDICT env var) as a hard stop in case the model still gets stuck. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> 2026-05-06 16:04:12 +02:00			`OLLAMA_NUM_PREDICT=4096`