cublas

Open repos and models their makers tagged cublas, most starred first. The same for everyone.

GitHubRepo

mmaemu

Bit-exact emulation of NVIDIA tensor-core FP8 GEMMs (torch._scaled_mm, Transformer Engine) as an H100 PCIe or RTX 4060 computes them, on the CPU or another CUDA GPU

  • 63
  • Python
  • Apache-2.0
  • 10 d ago
astdingTake it
AI on my own machineRepo

zcuda

zCUDA: 🔗 CUDA Binding Layer | ⚡ CUDA Kernel DSL. Write CUDA kernels in pure Zig. Or keep your C++ kernels. Ship them all to GPU.

  • 11
  • Zig
  • MIT
  • 9 mo ago
coderonionTake it
Search GitHub and Hugging Face for cublas

Get , open it once and sign in. Then come back here and press the button again.

cublas: open source repos and models on Take My Repo