What Ollama provides
Ollama runs the model. Disp8ch provides the workspace around it: WebChat, workflows, Data Sources, Boards, Council, Memory, and agent operations.
This makes local AI useful beyond a single chat box. A local model can summarize files, draft workflows, answer from notebooks, and create follow-up tasks.
Best first setup
Start Ollama, load a model that fits your computer, then choose Local AI in onboarding. Use the connection test before running workflows.
- Default Ollama base URL: http://127.0.0.1:11434
- Use Balanced recommendations when unsure.
- Keep web search and channel credentials optional.
When to use hosted models too
Use local models for private routine work. Add a hosted model only for jobs that need stronger planning, larger context, or specialized tool behavior.
Next step
This doc mirrors the in-app Docs tab. Install Disp8ch, open onboarding, and use WebChat when you want the app to inspect your current setup.