Pinned Loading

1. gpt-oss-20B gpt-oss-20B Public

   A PyTorch implementation of the GPT-OSS-20B architecture. All components are coded from scratch: RoPE with YaRN, RMSNorm, SwiGLU with clamping and residual connection, Mixture-of-Experts (MoE), Sel…

   Python 216 15

2. attn-arena attn-arena Public

   PyTorch implementations and benchmarks of hardware efficient attention variants for LLM inference on consumer single and multi GPU setups, with Flash Attention 2 backend.

   Python

3. h100_gemm h100_gemm Public

   A series of high-performance GEMM (General Matrix Multiply) implementations Iteratively optimised for H100 GPUs in Pure CUDA.

   Cuda 71 10

4. CUDA_Kernels CUDA_Kernels Public

   Random ML CUDA Kernels.

   Cuda

5. vllm vllm Public

   Forked from vllm-project/vllm

   A high-throughput and memory-efficient inference and serving engine for LLMs

   Python

6. flash-attention-2-triton flash-attention-2-triton Public

   Python