Popular repositories Loading

1. gemm3 gemm3 Public

   GEMM³: NVFP4 GEMM for NVIDIA B300 / GB300 in CuTe DSL

   Python 3

2. decon decon Public

   Forked from allenai/decon

   `decon`, but with python API binding.

   Rust 2

3. smem-swizzle-demo smem-swizzle-demo Public

   Shared-memory swizzling on Blackwell, measured: CuTe C++ & CuTe DSL benches (float4 + ldmatrix), ncu bank-conflict counters, CUPTI timing — 7.8x from one XOR on a B300

   Python 2

4. cs336-a1-sol cs336-a1-sol Public

   Speed-of-light study on the CS336 A1 45-min OWT leaderboard task: 2.9708 val@512 on B200 (best-ever entry: 3.0354). A measurement of the ceiling, not an entry.

   Python 1

5. vincentzed vincentzed Public

   GPU inference and model serving.

   1

6. vllm vllm Public

   Forked from vllm-project/vllm

   Python