Crimson AI NewsA CrimsonLingua Network service
EN ع
← Back to news
Research paper Hugging Face

FlashPrefill V2: Block-Sparse Attention Speeds Up Long-Context LLM Serving by Up to 47x

AI By Crimson AI Hugging Face Papers 21 August 2026 · 00:00 9 views
Share: X Telegram

Hugging Face researchers present FlashPrefill V2, an optimized sparse attention operator for the prefill phase of long-context LLMs, achieving up to 47.26x speedup over FlashAttention-2 on NVIDIA H20 GPUs.

FlashPrefill V2: Block-Sparse Attention Speeds Up Long-Context LLM Serving by Up to 47x

Key points

Long-context modeling is a key capability for large language models, but the quadratic complexity of attention remains a major bottleneck, especially during the compute-intensive prefill phase. FlashPrefill V2, introduced by Hugging Face researchers, evolves the earlier FlashPrefill prototype into a production-ready solution for long-context serving.

The new version introduces a mean correction term that suppresses approximation error, keeping performance degradation manageable even at extreme sparsity levels. It also redesigns the sparse attention operator with PackGQA memory access, warp specialization, and pingpong pipelining, aligning with the latest FlashAttention-3/4 implementations and supporting FP8 inference.

FlashPrefill V2 natively supports paged KV cache and continuous batching, allowing integration as an attention backend in modern inference frameworks such as SGLang. Extensive evaluations on NVIDIA H20 GPUs demonstrate significant speedups over dense baselines.

At 128K context length, FlashPrefill V2 achieves up to 47.26x and 27.19x speedups over FlashAttention-2 under FP8 and BF16 precision, respectively. In FP8, it still delivers a 30.49x speedup against an FA3/4-aligned dense baseline.

PrecisionBaselineSpeedup at 128K
FP8FlashAttention-247.26x
BF16FlashAttention-227.19x
FP8FA3/4-aligned dense30.49x
Source
Hugging Face · Hugging Face Papers
Related news
Research paper
Hugging Face 29 Aug 2026

Hugging Face Audit: 110 of 124 AI Evaluations Fail to Support Their Claims

A new commit-bound census of 124 Inspect Evals units reveals that 110 stop before deterministic inference due to missing historica...

4
Research paper
Hugging Face 29 Aug 2026

Aphanta: New Framework Diagnoses When Image Editing Boosts Multimodal Reasoning

Hugging Face researchers introduce Aphanta, a diagnostic framework that evaluates when image-editing intermediates improve multimo...

5
Research paper
Hugging Face 29 Aug 2026

EditaLive! Enables Real-Time Character Video Editing for Live Streaming

Hugging Face researchers introduce EditaLive, a framework for real-time human-centric video editing in live streams, achieving sta...

4