Popular repositories Loading

1. openfang openfang Public

   Open-source Agent Operating System

   Rust 18k 2.3k

2. picolm picolm Public

   Run a 1-billion parameter LLM on a $10 board with 256MB RAM

   C 1.7k 213

3. autokernel autokernel Public

   Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.

   Python 1.4k 146

4. qwen3.5-triton qwen3.5-triton Public

   Pure Triton kernels for Qwen3.5-27B inference on NVIDIA B200

   Python 117 9

5. rightnow-cli rightnow-cli Public

   Claude Code for CUDA. Free AI assistant that actually understands GPU architecture

   Python 109 22

6. AutoMegaKernel AutoMegaKernel Public

   An agent harness that compiles a model into one provably-correct, self-retargeting CUDA megakernel and self-tunes it past cuBLAS at batch-1 LLM decode, paper: https://arxiv.org/abs/2606.09682

   Python 90 9