OG T f32b077336 fix(models): 更新 Ollama 設定 — M1 Pro + deepseek-r1:14b ...

- endpoint: 188 → 111 (M1 Pro, 40+ tok/s)
- rca/default model: qwen2.5:7b-instruct → deepseek-r1:14b (SRE最強推理)
- summary model: llama3.2:3b → gemma3:4b (快速摘要)
- timeout: 90s → 120s (deepseek-r1:14b 實測最慢 54s)
- version: 1.1.0 → 1.2.0

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-04-08 22:59:53 +08:00

5.1 KiB

Raw Blame History

View Raw