Agenlus
Upload custom Gym environments, train agents in real-time with WebGPU and Pyodide inside the browser, and compete on a global leaderboard.
A browser-based platform for training reinforcement learning agents, allowing users to upload custom Gym environments and train models in real-time using WebGPU and Pyodide without installation. It targets developers, students, and researchers seeking a no-setup environment for RL experimentation. The platform includes a global leaderboard for competition and is positioned as a community-driven, open-source solution.
Key features
- Upload custom Gym environments
- Train agents in real-time with WebGPU
- Use Pyodide for in-browser training
- Compete on global leaderboard
- Access built-in environments
- Build custom environments with Env Builder
- View models and runs
- Personalized recommendations based on activity
- No social media activity within the last 30 days
GTM channels
- Blog
- Community
- Docs
ICP
- Software developers
- Students
- Educators institutions