Magnet:
- A local AI extension for VSCode designed to accelerate your development workflow.
Features:
How To Use:
VSCode Configuration Required:
Ensure that ollama was downloaded:
- MacOS && Linux:
curl -fsSL https://ollama.com/install.sh | sh
- Window:
irm https://ollama.com/install.ps1 | iex
To pull the corresponding ollama model for focusing on coding task only, I recommend using the qwen2.5-coder model;
- Depending on your device's hardware and performance capabilities, choose from the following models:
qwen2.5-coder:0.5b Pull Command (ollama pull qwen2.5-coder:0.5b)
qwen2.5-coder:1.5b Pull Command (ollama pull qwen2.5-coder:1.5b)
qwen2.5-coder:3b Pull Command (ollama pull qwen2.5-coder:3b)
qwen2.5-coder:7b Pull Command (ollama pull qwen2.5-coder:7b) (Recommend)
qwen2.5-coder:14b Pull Command (ollama pull qwen2.5-coder:14b)
qwen2.5-coder:32b Pull Command (ollama pull qwen2.5-coder:32b)
Ensure CodeLens is enabled in settings.json and make sure: "editor.codeLens": true
Ensure inLineSuggest is enabled in settings.json and make sure: "editor.inlineSuggest.enabled":true
License:
- Distributed under MIT license.
| |