PrototypeCommon
Infatoshi/MegaQwen
Qwen3-0.6B megakernel: 527 tok/s decode on RTX 3090 (3.8x faster than PyTorch)
- Owner
- Infatoshi (personal account)
- Stars
- Stars: 145
- Forks
- Forks: 10
- Language
- Cuda
- License
- None
- Created
- 2026
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by Infatoshi
All 8 repos by Infatoshi- Infatoshi/cuda-courseStars: 4,166CudaRC
- Infatoshi/fcc-intro-to-llmsStars: 846Jupyter NotebookBeta
- Infatoshi/mnist-cudaStars: 529CudaAlpha
- Infatoshi/x-cliStars: 431PythonAlpha
- Infatoshi/netheriteStars: 356CAlpha
- Infatoshi/all-of-itStars: 302Jupyter NotebookAlpha
Nearby in Cuda
All Cuda repos- lixincn2015/Partial-Order-PruningStars: 147CudaPrototype
- eunomia-bpf/basic-cuda-tutorialStars: 146CudaPrototype
- yalun753/moe-l2Stars: 146CudaPrototype
- enCRYPTON-labs/HEonGPUStars: 144CudaPrototype
- meder411/PyTorch-EMDLossStars: 143CudaPrototype
- leimao/CUTLASS-ExamplesStars: 143CudaPrototype