{"categories":[{"label":"Artificial Intelligence","url":"https://skillfed.io/packages/category/scientific-engineering-artificial-intelligence/3"}],"enrichment":{"capability":"GPU-accelerated attention mechanism implementation using CuTeDSL for Hopper and Blackwell GPUs, optimizing the core computational bottleneck in transformer models.","skillfed_tags":["gpu-kernels","transformer-optimization","cuda"],"use_cases":["Accelerate training or inference of large language models on Hopper/Blackwell GPUs by replacing standard attention with optimized kernels.","Reduce memory consumption during transformer model training by fusing attention operations into a single GPU kernel.","Implement causal attention for autoregressive language generation with lower latency.","Handle variable-length sequences efficiently in batched transformer inference without padding overhead."],"what_it_does":"flash-attn-4 is a specialized GPU kernel library that reimplements the attention operation\u2014a core component of transformer neural networks\u2014using CuTeDSL for modern NVIDIA Hopper and Blackwell architectures. It trades general-purpose compatibility for speed and memory efficiency on these specific GPUs by fusing multiple attention computation steps into a single kernel, reducing memory bandwidth and improving cache locality.\n\nThe package is designed for researchers and practitioners building or fine-tuning large language models and other transformer-based systems where attention computation dominates runtime. It provides two main entry points: flash_attn_func for standard attention and flash_attn_varlen_func for variable-length sequences. Installation requires torch, nvidia-cutlass-dsl, and a compatible CUDA version; the package is in active development (alpha status) and updated frequently.","worth_installing":"Yes, if you are training or running transformer models on Hopper or Blackwell GPUs and want lower latency and memory usage. The package is actively maintained, has no known vulnerabilities, and uses a permissive license. However, it is in alpha status and only useful for specific GPU hardware; it will not benefit users on other GPU architectures or CPU-only setups."},"id":"flash-attn-4","links":{"html":"https://skillfed.io/packages/flash-attn-4","md":"https://skillfed.io/packages/flash-attn-4.md","pypi":"https://pypi.org/project/flash-attn-4/"},"maintenance":{"status":"active"},"meta":{"latest_release":"2026-08-12","license_spdx":null,"license_treatment":"permissive","name":"flash-attn-4","python_support":"supports_current","summary":"Flash Attention CUTE (CUDA Template Engine) implementation"},"popularity":{"monthly_downloads":2040273,"position":3347,"tier":"top_5000"},"security":{"n_vulnerabilities":0},"version":"4.0.0b26"}