Skip to content

v1.0.0 – Local model (Ollama) + Roast AI

Latest
Latest

Choose a tag to compare

View all tags
@Alvaromp3 Alvaromp3 released this 21 Feb 03:52
· 102 commits to main since this release
v1.0.0
7fab75a

Release v1.0.0

Local model (Ollama)

Roast AI and multiverse support a local model via Ollama:

  • Vision (image + roast): OLLAMA_VISION_MODEL=llama3.2-vision:11b (e.g. ollama pull llama3.2-vision:11b)
  • Humor/roast text fallback: OLLAMA_HUMOR_MODEL=llama3.2 (e.g. ollama pull llama3.2)

Set in .env:

  • OLLAMA_BASE_URL=http://localhost:11434
  • OLLAMA_VISION_MODEL=llama3.2-vision:11b
  • OLLAMA_HUMOR_MODEL=llama3.2

If Ollama is not running or the request fails, the app falls back to OpenRouter (cloud).

Other changes

  • Multiverse / Roast AI with truth extraction and roast generation
  • Chat with image uploads and Roast tab in AI Assistant
  • OpenRouter vision model for chat when images are attached
Assets 2
Loading