Run Claude Code on local models via Ollama
A guide to setting up Claude Code in IntelliJ and connecting it to local models through Ollama for coding work.
Done withClaude Code
- What they did
- The author installed Claude Code with npm and ran it from IntelliJ's built-in terminal. They pulled the qwen3.5:9b model in Ollama and pointed Claude Code at Ollama's Anthropic-compatible local endpoint with environment variables. An ollama launch command is offered as a shortcut. As a test, they asked it to build a Java console task manager. The person approved each change.
- How it went
- Per the author, the agent planned the work, created the files, ran the program and fixed its own errors, and the app worked. The author rates local quality as good only for small tasks, and speed depends on your hardware.
- Worth knowing
- Speed depends on your hardware, and the author says the setup only works if your machine can handle the model. Start with a small project.
Try it yourself with Claude Code
Help me set up Claude Code to run against a local model through Ollama on my [Mac/Windows/Linux] machine, using [IntelliJ/my editor] as my IDE. Walk me through installing Ollama, pulling a suitable coding model for my hardware ([RAM/GPU]), and pointing Claude Code at it. You're done when Claude Code answers a simple request about a file in [my project folder] using the local model. Show me any config files or commands before you change or run them.
Want help getting this working? Join the Vibe Coding Academy ↗ Run by Alex Finn, who also runs this site.
Source: DEV · Oct 21, 2026
Related use cases
Turn tickets into merged pull requests
Pulls tasks from GitHub Issues or Linear, then runs Claude Code or Codex in isolated Kubernetes pods to carry each one through to a PR.
Extract vendor PDFs and create QuickBooks invoices
A custom integration built with Claude Code extracts data from vendor PDFs and creates categorized invoices in QuickBooks Online.
Write Playwright tests with Playwright MCP and Claude Code
A walkthrough of using Playwright MCP with Claude Code to generate Playwright tests in minutes.
Build a C compiler with a team of parallel agents
Anthropic's engineering write-up on tasking Opus 4.6 agent teams with building a C compiler.
Triage P2 bugs into draft PRs with demo videos
A Dot connected to GitHub and Slack watches #bugs, scopes and tests a fix in a cloud environment, and opens a draft PR with a short demo video.
Shepherd PRs to merge automatically
Bots check PRs every 30 minutes, send failures back to a cloud agent, review clean PRs and auto-merge high-confidence, low-risk ones.
Get 5 like this every morning
The best things people got an AI agent to do, each with the prompt to try it.