- Use larger coder model (32.8B params) for better reasoning
- Update client default model
- Update all AI routes to use qwen2.5-coder-8k:latest
- Already installed and available on server Ollama
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
- Point AI client to localhost:11434/v1 (Ollama)
- Remove DeepSeek API key dependency
- Update model name to qwen3:14b across all routes
- Update test mocks to use createLocalLLMClient
- Removes need for DEEPSEEK_API_KEY setting
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
- embedText: call Gemini REST API (768-dim) instead of Voyage (1024-dim)
- Migration: drop+recreate incidents.embedding as vector(768), update
match_incidents function, swap VOYAGE_API_KEY → GOOGLE_AI_API_KEY in app_settings
- Settings UI: relabel to "Google AI API Key (Embeddings)"
- All call sites updated (incidents POST, similar GET)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CPf5Rc8QPx6V8KLEEgfKEQ
- lib/claude/client.ts: replace Anthropic SDK with openai package pointed at DeepSeek baseURL
- 4 AI routes: port tool definitions, tool_choice, and output parsing to OpenAI function-calling format
- Drop thinking:{type:'adaptive'} (no DeepSeek equivalent); model string → deepseek-chat
- settings/route.ts: add DEEPSEEK_API_KEY to ALLOWED_KEYS
- migration: seed DEEPSEEK_API_KEY placeholder row in app_settings
- tests: update 3 AI route tests to mock createDeepSeekClient + OpenAI response shape
Voyage AI embedding path untouched (DeepSeek has no embeddings endpoint).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01CPf5Rc8QPx6V8KLEEgfKEQ