Improve LLM client immutability and CI model defaults.

Replace mutable Ollama model export with a const fallback and initializeModel return value, resolving the model from the environment after optional API discovery. Use a for-of loop over attempt indices instead of let in the retry path. Continue PDF generation when map image generation or upscaling fails, and avoid mutating request headers in place. Document Open WebUI-style URLs in the README, pin OLLAMA_MODEL in the Gitea release workflow, and adjust integration and unit tests for the new initialization behavior.
2026-04-14 22:42:21 -04:00
parent 4428cd4cb8
commit a6327fe573
7 changed files with 93 additions and 77 deletions
@@ -15,6 +15,7 @@ jobs:
    env:
      OLLAMA_API_URL: https://ai.keligrubb.com/api/chat/completions
      OLLAMA_API_KEY: ${{ secrets.OLLAMA_API_KEY }}
+      OLLAMA_MODEL: qwen3.5-122b-a10b
      COMFYUI_URL: http://192.168.1.124:8188
    steps:
      - name: Checkout