skills.homes logoskills.homescapability registry
home/categories/framework-internals/math-inc-opengauss-skills-mlops-training-flash-attention-skill-md
cd ..
framework-internalsdevelopment

optimizing-attention-flash

Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Use when training/running transformers with long sequences (>512 tokens), encountering GPU memory issues with attention, or need faster inference. Supports PyTorch native SDPA, flash-attn library, H100 FP8, and sliding window attention.

Voir le code sourceframework-internals
math-inc
maintainer
math-inc
Mis à jour 3/19/2026
Étoiles
1165
Forks
96
quick start

Installation and usage

Optimizes transformer attention with Flash Attention for 2-4x speedup and 10-20x memory reduction. Use when training/running transformers with long sequences (>512 tokens), encountering GPU memory issues with attention, or need faster inference. Supports PyTorch native SDPA, flash-attn library, H100 FP8, and sliding window attention.

Installation
$ install --globalskills.sh
Utilisation

Après l'installation, vous pouvez utiliser ce skill en exécutant la commande suivante dans votre terminal :

skills use optimizing-attention-flash