LTSLegendary
Dao-AILab/flash-attention
Fast and memory-efficient exact attention
- Owner
- Dao-AILab (organization)
- Stars
- Stars: 25,245
- Forks
- Forks: 3,204
- Language
- Python
- License
- BSD-3-Clause
- Created
- 2022
- Tier
- LTS (20,000 ★ and up) What this tier means
More repos by Dao-AILab
All 7 repos by Dao-AILab- Dao-AILab/quackStars: 1,189PythonBeta
- Dao-AILab/causal-conv1dStars: 975PythonBeta
- Dao-AILab/sonic-moeStars: 787PythonBeta
- Dao-AILab/fast-hadamard-transformStars: 357CAlpha
- Dao-AILab/gram-newton-schulzStars: 189PythonPrototype
- Dao-AILab/grouped-latent-attentionStars: 139PythonPrototype
Nearby in Python
All Python repos- earthtojake/text-to-cadStars: 25,297PythonLTS
- deepseek-ai/DeepInferStars: 25,291PythonLTS
- junyanz/pytorch-CycleGAN-and-pix2pixStars: 25,276PythonLTS
- cookiecutter/cookiecutterStars: 25,174PythonLTS
- m-bain/whisperXStars: 25,081PythonLTS
- haotian-liu/LLaVAStars: 25,077PythonLTS