Use Claude Code for Free and Without Limits, Running on Your Own Machine

Description and use

Developers can use Claude Code for free by running a local Ollama server, eliminating per‑request costs and keeping code private.

Problem

Users face usage‑based pricing when using Claude Code, limiting frequent or large‑scale use.

Outcome

By redirecting Claude Code to a local Ollama server, users eliminate per‑request costs, keep code on‑premises, and gain offline capability, though model performance depends on local hardware.

Workflow

• Install Ollama on your machine.
• Pull the desired Claude model (e.g., claude‑3‑sonnet).
• Run ollama launch claude or ollama launch claude --model <name> to start the local server.
• Set environment variables: CLAUDE_SERVER_ADDRESS=localhost, CLAUDE_SERVER_PORT=11434, CLAUDE_TOKEN=ollama, CLAUDE_KEY= (empty).
• Configure Claude Code to point to the local server address.
• Use Claude Code as usual; requests are handled locally.
• Monitor performance and adjust model size based on available RAM (32 GB+ for 20‑30B models, 7‑14B for standard hardware).

Evidence

Claude Code talks to a model through environment variables that define a server address. Point that address at a local server instead of the company's, and every request goes to your machine.

Source

Install Ollama, pull the model you want, then run one command: ollama launch claude and pick your model, or pass it directly with ollama launch claude --model <name>.

Source

The token variable is where people get it wrong. Set it empty and the connection fails. It has to be the word ollama.

Source