The #1 FREE AI-powered Git commit message generator for VS Code
Generate detailed, professional, conventional commit messages in seconds using GPT-5, Claude 4, Gemini 2.5 Pro, DeepSeek, Llama 3.3, and 15+ more AI models.
Tired of writing commit messages? Commit Genius generates detailed, professional, conventional commit messages automatically using the latest AI models — completely FREE.
- 🆓 100% Free to Start: Use powerful free AI models with no credit card required
- ⚡ Instant Generation: Create perfect commit messages in under 4 seconds
- 🎯 Smart Staging: Automatically stages your changes - no manual work needed
- 📊 Conventional Standards: Follows industry-standard commit conventions (feat, fix, chore, etc.)
- 🔒 Enterprise Security: Your code never leaves your machine - only diffs are analyzed
- ⚡ Lightning Fast: Generate commit messages in under 4 seconds
- 🤖 Multiple AI Models: OpenAI GPT, Claude Sonnet, Google Gemini, DeepSeek, Mistral, Qwen
- 📏 Conventional Commits: Perfect
feat:,fix:,chore:,docs:formatting every time - 🆓 Free Forever: Start with powerful free models, upgrade when you need more
- 🔒 Bank-Level Security: API keys encrypted in VS Code's secure storage
- 🎨 Zero Setup: Works instantly with your existing Git workflow
- 📈 Productivity Boost: Save 5+ minutes per commit, focus on coding instead
- Generate commit messages using state-of-the-art AI models
- Support for both free and premium AI models
- Intelligent analysis of your staged changes
- Automatically formats messages with proper prefixes (
feat:,fix:,chore:, etc...) - Maintains consistent commit history
- Improves project maintainability and readability
- One-click generation via sparkle button in Source Control
- Seamless VS Code integration
- Edit generated messages before committing
- Secure API key management
- Free Models: DeepSeek, Qwen, Mistral, OpenAI GPT-OSS
- Premium Models: Claude Sonnet 4, Gemini 2.5 Pro, Grok Code Fast
- Easy model switching through VS Code settings
Switch between free and premium AI models in settings
Click the sparkle button in Source Control to generate AI commit messages
- Open Visual Studio Code
- Press
Ctrl+Shift+X(Windows/Linux) orCmd+Shift+X(Mac) to open Extensions - Search for "Commit Genius"
- Click Install on the extension by
aicommit-publisher - Reload VS Code if prompted
This extension is also available on the Open VSX Registry for users of alternative IDEs like Trae, Kiro, VSCodium, and other VS Code compatible editors:
- Open your IDE's extension panel
- Search for "Commit Genius" or "AI Git Commit"
- Install the extension from the marketplace
- Restart your IDE if needed
Look for the ✨ sparkle button in your Source Control panel - you're ready to go!
- Visit OpenRouter.ai
- Sign in with GitHub or Google
- Click "Create Key"
- Copy your API key (starts with
sk-or-...)
- Make some changes in your Git repository
- Click the ✨ sparkle button in the Source Control panel (staging happens automatically)
- Paste your API key when prompted
- Watch as your commit message is generated!
- Go to Settings → Extensions → Commit Genius
- Choose your preferred AI model
- Adjust settings to match your workflow
| Model | Provider | Description |
| ----------------------------------------------- | ------------ | -------------------------------------- | ---- | ------------------------- | -------- | ------------------------------ |
| deepseek/deepseek-chat-v3-0324:free | DeepSeek | Advanced coding model (71B parameters) |
| qwen/qwen3-coder:free | Qwen | Specialized code understanding |
| mistralai/mistral-small-3.2-24b-instruct:free | Mistral | Balanced performance and speed |
| openai/gpt-oss-120b:free | OpenAI | Large context understanding |
| meta-llama/llama-3.3-70b-instruct:free | Meta Llama | multilingual large language model |
| nvidia/llama-3.1-nemotron-ultra-253b-v1:free | NVIDIA | Advanced reasoning model |
| deepseek/deepseek-r1-0528:free | DeepSeek | Advanced reasoning model |
| xiaomi/mimo-v2-flash:free | Xiaomi | Fast and efficient coding assistant |
| google/gemma-3-27b-it:free | Google | Instruction-tuned language model |
| nousresearch/hermes-3-llama-3.1-405b:free | NousResearch | Large 405B parameter model |
| arcee-ai/trinity-mini:free | Arcee AI | Compact high-performance model | \r\n | allenai/molmo-2-8b:free | Allen AI | Multimodal open language model |
| Model | Provider | Description |
|---|---|---|
anthropic/claude-sonnet-4 |
Anthropic | Superior reasoning and code analysis |
google/gemini-2.5-pro |
Advanced multimodal capabilities | |
x-ai/grok-code-fast-1 |
X.AI | Fast and efficient code generation |
deepseek/deepseek-chat-v3.1 |
DeepSeek | Enhanced version with better accuracy |
mistralai/mistral-7b-instruct |
Mistral | High-quality responses |
qwen/qwen3-30b-a3b-instruct-2507 |
Qwen | Advanced reasoning model |
openai/gpt-5 |
OpenAI | Cutting-edge language understanding |
{
"commit-genius.model": "openai/gpt-oss-120b:free"
}- Keep changes focused: Make focused, related changes for more accurate commit messages
- Review before committing: Always review the generated message and edit if needed
- Use descriptive file names: Better file names lead to better commit messages
- Commit frequently: Smaller, focused commits generate better messages
If you see an error like:
This happens because most free models on OpenRouter do not support Zero Data Retention (ZDR).
When ZDR is enabled, free endpoints are filtered out, which causes the 404 error.
- Go to OpenRouter Privacy Settings.
- Under Training, Logging & Privacy, configure your settings as shown below:
- Enable Free endpoints that may train on inputs ✅
- Enable Free endpoints that may publish prompts ✅
- Enable Input/output logging for all requests ✅
- Disable ZDR Endpoints Only ❌ (turn this off if you want to use free models)
Example configuration:
- If you require ZDR for compliance, you’ll need to use paid models (e.g., Claude, GPT-4 Turbo, Gemini Pro) because free models don’t support ZDR.
- For most personal projects, the above settings will allow free models to work smoothly.
- Free models have usage limits
- If you hit a rate limit, wait a few minutes or switch to a other free models or premium models
- Consider upgrading to premium models for heavy usage
- Ensure your API key starts with
sk-or- - Check that your OpenRouter account has sufficient credits if using paid model or continue with free models
- Re-enter your API key if authentication fails
- Ensure you're in a valid Git repository
- Make sure you have changes in your working directory (staging happens automatically)
- Check that Git is properly installed and configured
+ Added user authentication middleware
+ Updated package.json version
- Removed deprecated API endpointsfeat(auth): add user authentication middleware
- Implement JWT-based authentication middleware for API routes
- Update package.json version to reflect new auth capabilities
- Remove deprecated API endpoints that were replaced by new auth system
This change improves security by requiring authentication
for protected routes and cleans up legacy code.
| Change Type | Generated Message |
|---|---|
| Bug fix | fix(api): resolve null pointer exception in user service |
| New feature | feat(ui): implement dark mode toggle with system preference detection |
| Documentation | docs(api): update REST API documentation with authentication examples |
| Refactoring | refactor(db): optimize database queries reducing response time by 40% |
| Dependencies | chore(deps): update dependencies to latest stable versions |
- Local Storage: API keys are stored securely in VS Code's encrypted secret storage
- No Data Collection: We don't collect or store your code or commit messages
- Secure Transmission: All API requests use HTTPS encryption
- No Telemetry: No usage data is sent to external servers
- Keys are never logged or exposed in plain text
- Stored using VS Code's built-in secret management
- Only transmitted to OpenRouter.ai for API requests
- Can be updated or removed at any time
This project is licensed under the MIT License - see the LICENSE file for details.
If you find Commit Genius helpful, please:
- ⭐ Star the repository on GitHub
- 📝 Leave a review on the VS Code Marketplace
A: Yes! Commit Genius offers 10+ completely FREE AI models including DeepSeek, Qwen, Mistral, and Llama. Premium models like GPT-5 and Claude 4 are available for users who want cutting-edge performance.
A: Commit Genius is a FREE alternative to GitHub Copilot for commit messages. While Copilot requires a $10/month subscription, Commit Genius offers free AI models with no subscription required. It's specifically optimized for generating detailed, conventional commit messages.
A: Commit Genius generates detailed, multi-line commit messages following conventional commit standards. It analyzes your actual code changes and creates professional messages with subject lines, body descriptions, and proper formatting — saving you 5+ minutes per commit.
A: 15+ models including OpenAI GPT-5, Claude Sonnet 4, Google Gemini 2.5 Pro, DeepSeek, Mistral, Qwen, Llama 3.3, and more via OpenRouter.ai.
A: Yes! Commit Genius works with VS Code and all compatible editors including Cursor, Windsurf, VSCodium, Kiro, and Trae via the Open VSX Registry.
ai commit message generator git commit ai conventional commits auto commit message gpt commit claude commit gemini commit free ai extension copilot alternative free git automation commit helper semantic commits ai git assistant vscode git extension commit message writer deepseek llama openai vscode source control ai
Made with ❤️ by Anmol