Repositories list

• triton-kernels

  Public
  Triton kernels for Flux

  Python

  •

  MIT License

  •0•18•0•0•Updated Jan 1, 2025Jan 1, 2025

• flux

  Public
  Official inference repo for FLUX.1 models

  Python

  •

  Apache License 2.0

  •1.4k•8•0•1•Updated Jan 1, 2025Jan 1, 2025

• test_attn

  Public
  Testing and benchmarking different attention implementations and backends

  Python

  •2•5•0•0•Updated Dec 5, 2024Dec 5, 2024

• flux-tinygrad-opt

  Public
  Optimize Flux on tinygrad

  Python

  •

  MIT License

  •1•0•0•0•Updated Oct 18, 2024Oct 18, 2024

• flash-attention

  Public
  Fast and memory-efficient exact attention

  Python

  •

  BSD 3-Clause "New" or "Revised" License

  •1.4k•0•0•0•Updated Oct 10, 2024Oct 10, 2024

• nexfort

  Public
  OneDiff compiler infrastructure using torch Inductor

  Python

  •0•1•0•0•Updated Sep 26, 2024Sep 26, 2024

• kernels

  Public
  Python

  •19•0•0•0•Updated Sep 9, 2024Sep 9, 2024

• xDiT

  Public
  xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) on multi-GPU Clusters

  Python

  •

  Apache License 2.0

  •97•0•0•0•Updated Aug 30, 2024Aug 30, 2024

• quanto

  Public
  Python

  •0•1•1•0•Updated Aug 14, 2024Aug 14, 2024

• minRF-ONNX

  Public
  Minimal implementation of scalable rectified flow transformers, based on SD3's approach

  Jupyter Notebook

  •

  Apache License 2.0

  •41•5•0•0•Updated Jul 27, 2024Jul 27, 2024

• quant_dit_models

  Public
  Python

  •0•0•0•0•Updated Jul 25, 2024Jul 25, 2024

• cutlass-kernels

  Public
  Cuda

  •

  MIT License

  •30•0•0•0•Updated Jul 11, 2024Jul 11, 2024

• unet.cu

  Public
  UNet diffusion model in pure CUDA

  Cuda

  •27•0•0•0•Updated Jun 28, 2024Jun 28, 2024