A coding agent that runs a local model on your Mac.
No cloud inference, no API key, no account.
localcode runs an open-weight model on your Mac and uses it to read, edit and test your code. Your prompts and your files stay on your machine. The only thing it downloads is the model weights, once per model.
pip install -U localcode # or: uv pip install -U localcodeThe inference server ships inside the package. Nothing is compiled or cloned on your machine.
cd your-project
localcodeOn first launch localcode recommends a model for your Mac's memory. Pick one, wait for the download, and start typing.
> Implement the retry decorator in retry.py so every test in test_retry.py passes. Then run: pytest -q
Docs: mjwsolo.github.io/localcode
- Reads and edits files in your project
- Runs your tests, builds, Git and shell commands, and asks before anything risky
- Searches code by name, content or structure
- Scaffolds and launches apps, then checks that they respond
- Remembers the task across messages
- Mac with Apple Silicon
- 16 GB unified memory or more
- Python 3.10 or newer
- About 12 GB of free disk for the smallest model
localcode recommends a model by your Mac's memory and marks it with a star. You choose; nothing is selected for you. Every model runs on binaries shipped in the package.
| Model | Weights | Quant | Active params | Min RAM |
|---|---|---|---|---|
| Gemma 4 12B | 7.4 GB | UD-Q4_K_XL | 12B (dense) | 16 GB |
| Qwen 3.6 35B-A3B | 10.7 GB | UD-IQ2_M | 3.0B (MoE) | 24 GB |
| Gemma 4 26B-A4B | 11.2 GB | UD-IQ3_S | 3.8B (MoE) | 24 GB |
| DiffusionGemma 26B-A4B | 15.7 GB | Q4_K_M | 4B (diffusion MoE) | 32 GB |
| Muse Glimmer 30B | 15.9 GB | UD-Q4_K_XL | 30B (dense, vision) | 32 GB |
| Qwen 3.8 27B | 17.9 GB | UD-Q4_K_XL | 27B (dense) | 36 GB |
| North-Mini-Code 30B-A3B | 17.9 GB | UD-Q4_K_M | 3B (MoE) | 36 GB |
| Gemma 4 12B (full) | 23.8 GB | BF16 | 12B (dense) | 48 GB |
| Gemma 4 26B-A4B | 28.0 GB | UD-Q8_K_XL | 3.8B (MoE) | 64 GB |
| Qwen 3.6 35B-A3B | 38.5 GB | UD-Q8_K_XL | 3.0B (MoE) | 96 GB |
Min RAM is the memory at which localcode will recommend the model. You can pick a heavier one by hand. DiffusionGemma is a research model that is never recommended automatically.
Measured on a MacBook Pro (M5 Max, 128 GB) with Qwen 3.6 35B-A3B UD-IQ2_M at a 131072-token context: about 89 tokens/s generation, about 1174 tokens/s prompt processing, and 12 to 15 seconds for a typical four-tool-call task.
Inference is local. Three features use the network: model downloads, the web_search and web_fetch tools, and any MCP servers you add. See Network Boundary for the full list.
Powerful, personal AI should work for everyone, on any device, anywhere. That means running it locally. localcode is a first step.
To sponsor localcode, reach out.
See CONTRIBUTING.md.
Apache 2.0. See LICENSE.