Download Oikyo Desktop

Public Beta - Latest Release: v0.3.39

Oikyo is a zero-setup ML platform for fine-tuning language models on your hardware. Free during beta. Download, install, and start training in minutes.

Windows

Download: Oikyo-0.3.39-win32-x64.exe

Requirements: Python 3.10+ (installed and available in PATH)

Installation:

1. Download the .exe installer
2. Double-click to run the installer
3. Follow the installation wizard
4. Launch Oikyo from Start Menu or Desktop

System Requirements:

• Windows 10 or 11 (64-bit)
• 4 GB RAM (8 GB recommended)
• 2 GB free disk space

macOS

Download: Oikyo-0.3.39-darwin-arm64.dmg

Requirements: Python 3.10+ (installed and available in PATH)

Installation:

1. Download the .dmg file
2. Open the downloaded DMG
3. Drag Oikyo to your Applications folder
4. Launch Oikyo from Applications
5. If macOS blocks the app (Gatekeeper):
   • Right-click on Oikyo → Select “Open”
   • Click “Open” in the security dialog
   • Subsequent launches will work normally

System Requirements:

• macOS 11 (Big Sur) or later
• Apple Silicon (M1/M2/M3) or Intel
• 4 GB RAM (8 GB recommended)
• 2 GB free disk space

Note: Oikyo automatically optimizes for Apple Silicon Macs using MLX for faster inference.

Linux

Download: Download all .partXX files from the releases page

Requirements: Python 3.10+, FUSE (usually pre-installed on modern Linux distributions)

Installation:

# 1. Download all part files (partaa, partab, etc.)
# You can use wget or download from the releases page

# 2. Reassemble the AppImage
cat Oikyo-*-linux-x86_64.AppImage.part* > Oikyo-linux-x86_64.AppImage

# 3. Make it executable
chmod +x Oikyo-linux-x86_64.AppImage

# 4. Run Oikyo
./Oikyo-linux-x86_64.AppImage

If FUSE is not installed:

# Ubuntu/Debian
sudo apt install libfuse2

# Fedora
sudo dnf install fuse

# Arch Linux
sudo pacman -S fuse2

System Requirements:

• Linux with kernel 3.10+ (Ubuntu 18.04+, Fedora 30+, etc.)
• 4 GB RAM (8 GB recommended)
• 5 GB free disk space

Getting Started

After installation:

1. Launch Oikyo - The app will initialize on first run
2. Sign in - Enter your email for a one-time login code, or sign in with Google/GitHub
3. Create your first environment - Choose a pre-configured template (LLM Training, Evaluation, or Inference)
4. Start training models - Browse the model library and begin fine-tuning!

Getting Started Guide: docs.oikyo.ai Community Discord: Join our Discord Support: GitHub Issues

Release Notes

See GitHub Releases for detailed changelog and release notes.