Ollama — run the AI locally (no cloud, no cost)

The Deputy and Posse can run on a LOCAL model via Ollama instead of the Anthropic cloud — pick the provider in the Deputy's "AI" dropdown (Anthropic (cloud), OpenAI / compatible, Google Gemini, or Ollama (local)). Local means: no API key, no per-token cost (live Deputy and Posse runs still need Pro or the trial), and nothing leaves your machine — when Ollama runs on this machine (the default host, localhost). Point the host at another machine and your prompts go there: the Deputy then gives the same off-machine WARNING as a cloud AI. Good for private, air-gapped, or just-experimenting work.

Setup:

1. Install Ollama (ollama.com) and make sure it's running — it serves on localhost:11434 by default (you can also point at another machine on your LAN).

2. Pull a TOOL-CAPABLE model: `ollama pull llama3.1` (also good: qwen2.5, mistral). Tool use — the whole point of the Deputy — needs a model that supports function calling; not every local model does.

3. In the Deputy: set AI = Ollama (local), confirm the host, and pick the model (the ↻ button lists installed ones).

How it works: SubnetSlinger talks to Ollama's /api/chat endpoint and translates to/from the same internal shape the Anthropic client uses — so the agent loop, all the read-only tools, and any connected MCP servers work identically. The token line shows in/out and says 'local (free)'. The Posse follows whatever provider the Deputy is set to.

Tradeoffs: local models are private and free but generally weaker at long multi-step tool chains and slower on modest hardware than the frontier cloud models. Use Haiku/Sonnet/Opus for the hardest investigations; use Ollama when privacy, cost, or offline operation matters more. Same safety either way — the built-in tools never change device state, and an MCP server can only if you tick its 'writes' box.

Does it get better over time? No — a local model's weights are FIXED, so it does not learn from your usage or build up better predictions the more you use it. Within a chat it sees the conversation so far (the context window), which resets on New chat. What does carry over is what you save in the Deputy's Project knowledge — instructions, Memory and Skills are sent with every chat in that project, to whichever provider you use. To get stronger answers you pull a bigger/newer model or switch to the cloud for that one hard investigation — not by using it more.

Two practical notes: the Ollama app must be RUNNING before you pick Ollama (start it first — the Deputy reaches it at localhost:11434), and the very first reply after a cold start can be slow while the model loads into memory; it speeds up once it's warm. It works — just give that first prompt a moment.

Under the hood this is a provider seam (IChatProvider): ClaudeClient, OllamaClient, OpenAiCompatClient, and GeminiClient are all implementations behind the same interface — which is why the Deputy, Posse, tools, and MCP behave identically whichever backend you pick.