Who it is forA developer who wants to try a coding agent on open models.
The problemWiring agents to local models by hand is error-prone.
The outcomeA coding agent session running through Ollama.
What you need first
- macOS, Windows, Linux or Docker
- Enough RAM or VRAM for the model you pick
Steps
- Install Ollama
- Run
ollama launch claude - Pick a model when prompted
- Work in your repo
Prompt for your agent
Run `ollama launch claude` and complete <small coding task> using an Ollama model. Report which model was used.
Caveat. Small local models are much weaker at agentic coding than frontier models.
Use this in your AI agent
Ask your agent
Using the Deduped MCP, get the setup steps for ollama/ollama and "Launch a coding agent backed by Ollama"
Add Deduped to Claude Code once
claude mcp add --transport http deduped https://dedupedai.com/mcp
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer