New from @Meta Engineering: FlashAttention-4 extended with MXFP8 support for @nvidia…
This is a AI post classified by Jev as AI infra & evals (a launch), kept by the AI Radar because it carries real work, not commentary.
New from @Meta Engineering: FlashAttention-4 extended with MXFP8 support for @nvidia Blackwell—from forward and backward kernels to fused quantization and jagged cross-attention. The team developed an end-to-end jagged module with fused quantization, FP8 activation and compute which is being used internally at Meta for GEM training. On the latest gen hardware, LP FA4 kernel reaches 2.85 PFLOP/s forward and 2 PFLOP/s backward, with up to a 1.30× end-to-end module speedup. ✍️ Devashish Shankar, Santosh Mohan, Jiaqi Xu, Darren Liu, Han Xu Explore the design and open source implementation in
Posted by PyTorch (512.7k followers) 2 days ago · 81 likes · 16.1k views · view the original post on X. Kept by the AI Radar as AI infra & evals.
More AI work like this
- Trains your own large language model from scratch using plain PyTorch. — @tom_doerr
- Here's a useful use-case for @typesafeai for a change. — @iam_zachi
- This is genius. — @daniel_mac8
- Which GPU should you use for embedding workloads? — @runpod
- Big update to the @Alibaba_Qwen Qwen3.8-Flash-Next single DGX Spark recipe! — @jvr0x
- 🔰 The Databricks Advanced Learning Festival runs September 16 through October 14, with… — @databricks
- Q: What is a trace? — @HamelHusain
- Inference scaling part 1. — @rasbt
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 27.2k posts from 4.8k X accounts over the last 14 days, 1.2k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-19 18:08 UTC. Full method.