Life & fun · September 20, 2024
LeanRL: Fast PyTorch RL with Torch.compile and CUDA Graphs
We're excited to announce that we've open-sourced LeanRL, a lightweight PyTorch reinforcement learning library that provides recipes for fast RL training using torch.compile and CUDA graphs. By leveraging these tools, we've achieved significant speed-ups compared to the original CleanRL implementations - up to 6x faster! Reinforcement learning is notoriously CPU-bound due to the high frequency of small CPU operations. PyTorch's powerful compiler can help alleviate these issues, but comes with its own costs. LeanRL addresses this challenge by providing simple recipes to accelerate your…
LeanRL: Fast PyTorch RL with Torch.compile and CUDA Graphs launched on September 20, 2024 with 53 votes, #597 of 5,281 launches that month and more than 90% of that year's launches. The site was still online in September 2026, 2 years on.
In plain words
LeanRL is a lightweight PyTorch reinforcement learning library offering recipes for accelerated RL training using torch.compile and CUDA graphs. It addresses the CPU bottleneck common in reinforcement learning by providing simple optimization techniques that achieve significant speedups—up to 6.8x for PPO on Atari, 5.7x for SAC, and 3.4x for TD3. Designed for researchers and practitioners, LeanRL provides single-file algorithm implementations with minimal dependencies, enabling faster experimentation and training loop optimization.
written from the facts on this page · September 2026
From the sources
LeanRL is a fork of CleanRL, where selected PyTorch scripts optimized for performance using compile and cudagraphs. - meta-pytorch/LeanRL
In the maker’s words, at launch
We're excited to announce that we've open-sourced LeanRL, a lightweight PyTorch reinforcement learning library that provides recipes for fast RL training using torch.compile and CUDA graphs. By leveraging these tools, we've achieved significant speed-ups compared to the original CleanRL implementations - up to 6x…
Does a similar job
all alternatives →
Terminal-Bench-RL: Training long-horizon terminal agents with RL2025 · github.com · ▲125After training calculator agent via RL, I really wanted to go bigger! So I built RL infrastructure for training long-horizon terminal/coding agents that scales from 2x A100s to 32x H100s (~$1M worth of compute!) Without any training, my 32B agent hit #19 on Terminal-Bench leaderboard, beating Stanford's Terminus-Qwen3-235B-A22! With training... well, too expensive, but I bet the results would be good! *What I did*: - Created a Claude Code-inspired agent (system msg + tools) - Built Docker-isolated GRPO training where each rollout gets its own container - Developed a multi-agent…
Rlpyt – Modular, Optimized Deep Reinforcement Learning in PyTorch2019 · github.com · ▲7Reinforcement Learning in PyTorch. Contribute to astooke/rlpyt development by creating an account on GitHub.
Supercharging RL with Hyper-Efficient Online Opt, +165% in 2h, $102025 · arc.computer · ▲32Arc enables continual learning for agents to adapt, learn, and transfer knowledge in live environments. Open-source SDK for building agents that learn from experience and improve over time.
I RL-trained an agent that trains models with RL (for ~$1.3k)Jul 2026 · github.com · ▲107RL-training an AI agent to RL-train AI agents. Contribute to Danau5tin/ai-trains-ai development by creating an account on GitHub.
Zipslicer, a library for loading LLM checkpoints on consumer hardware2023 · github.com · ▲38This is a low-level opensource library I developed for my own use and decided to share, as it makes it possible to process large checkpoints of neural networks without renting high-RAM instances, on a regular PC. It replaces torch.load() with a custom function that produces a dictionary that materializes tensors on the fly. Compared to other solutions it doesn't require sharding or re-encoding checkpoints and uses them completely as-is. It is a foundation to make it possible to run inference and compress language models and other large models one layer at a time - in principle, even one…
More life & fun this month
the category →
A free city-building game in your browser. You are the mayor. You know what the city needs. Now get it past the zoning board, the neighbors, the appeals court and the next election.
Life & fun · 20h ago · permit.petercai.com
- IV
Life & fun · 13d ago · tanzaniaeducationproject.org
Turn a phone, tablet or TV into a sign. Type a message and it fills the screen: welcome signs, timers, Wi-Fi passwords and more. No app, no account, free.
Life & fun · 14d ago · bigwords.page
Play a free virtual guitar online with your keyboard or by touch. Slow songs down and see every chord, string and strum. Acoustic, classic, electric and bass.
Life & fun · 8d ago · guitoro.com
Carrier settings from iPhone, Pixel and Galaxy firmware, decoded and compared: APNs, VoLTE, 5G and Wi-Fi Calling per carrier, and what each build changed.
Life & fun · 12d ago · carrierexplode.com
Life & fun · 22d ago · space.bl2.net
Source: Show HN post ↗ · launch text quoted in part
Launched alongside, September 2024
the whole month →
BeforeSunset AI 2.0▲1,267Personalized AI daily planning that suits your life
AI · 2024 · beforesunset.ai
