What it does
This n8n workflow runs on a local Ollama setup and picks the most suitable model for each incoming prompt. Text questions go to general models, programming requests to a code model and image questions to vision models. Conversation memory is kept and everything is processed locally, so data stays on your machine.
Use cases
- 01Route coding questions to a code-specific local model
- 02Send image questions to a vision-capable local model
- 03Keep sensitive prompts on local infrastructure
Connects
AI Agent · Ollama Chat Model · Simple Memory