autoresearch-mlx is a tool that runs autonomous AI research loops on Apple Silicon Macs. It lets you explore AI models without needing complex tools like PyTorch. This software is designed to run smoothly on Mac computers with Apple chips, making it easier to experiment with AI research without deep technical skills.
This version is built for Apple Silicon (M1, M2, and newer). It focuses on simplicity, speed, and independence from large AI libraries.
To run autoresearch-mlx, make sure your setup fits these needs:
- Operating System: macOS Big Sur or later (run on Apple Silicon chips only)
- Processor: Apple M1, M2, or newer
- Memory: At least 8 GB of RAM recommended
- Disk Space: Around 500 MB free space for the program and data
- Internet: Required for initially downloading the application and optional updates
This app is optimized for Mac and will not run on Windows or other platforms.
You do not need programming knowledge to start using autoresearch-mlx. Follow these steps carefully.
Click the big green button at the top or go directly to:
This page hosts the latest version and detailed information.
On the GitHub page, look for the Releases section.
- Click on the latest release version number (for example: v1.0).
- Look for a file designed for Apple Silicon Macs. This will usually have a
.dmgor.zipextension.
- Click the file name to start downloading.
- Depending on your browser, it may save the file in your Downloads folder.
- Open the downloaded
.dmgfile by double-clicking it. - A window will open showing the application icon.
- Drag the application icon to your Applications folder.
- Go to your Applications folder.
- Double-click autoresearch-mlx to start it.
- If macOS shows a warning about an unidentified developer, click Open Anyway in System Preferences > Security & Privacy.
autoresearch-mlx runs research loops on your Mac. It studies AI models over many iterations and adjusts itself using autonomous steps. This can help you learn about AI without needing to write code.
The app uses native macOS features and Apple’s optimized processing units. This approach speeds up AI tasks without heavy software like PyTorch.
When you open the app, you will see a simple interface.
- Start a Research Loop: Click the Start button to begin.
- Monitor Progress: The app shows a status screen with ongoing research details.
- Stop or Pause: Use the Stop or Pause button to control the process.
- View Results: After completion, results display clearly. You can explore what the model learned.
No setup or coding is needed for these steps.
If you want to customize how the app runs:
- Click on the Settings icon.
- Adjust options like the number of loops, speed, or data sources.
- Save your changes before returning to the main screen.
These options let you control how much work your Mac does during research.
autoresearch-mlx stores its files on your Mac.
- By default, data saves in your Documents/autoresearch-mlx folder.
- You can change this folder in Settings > Data Location.
- Make sure you have enough disk space if you run many loops.
To get the latest improvements:
-
Visit the download link again:
-
Go to the Releases section.
-
Download the newest version.
-
Replace your old application file by dragging the new one into Applications.
No. autoresearch-mlx runs only on Apple Silicon Macs with macOS Big Sur or later.
No. autoresearch-mlx works without extra software or coding libraries.
Yes. Use the Stop button on the main screen at any time.
Use the Issues tab on the GitHub page to submit bug reports or feature suggestions.
Start by visiting this page to download the latest version:
Click Releases, pick the latest .dmg or .zip file, download it, and follow the installation steps above.