# Zyphra's ZYFA cuts softmax work per score from 14 to 6 cycles

> Original title: FP8 makes matrix math 2x faster than BF16 but does not speed up softmax, so softmax becomes the bottleneck in long-context prefill.

- Source: Zyphra
- Published: 2026-11-06T02:36:47.000Z
- Bad Influence: https://badinfluence.news/items/l0r0wr5fe3zk1bz9b6h253wr1
- Original: https://x.com/ZyphraAI/status/2118533020345074092

## Summary

Zyphra says its ZYFA "lean P" replaces softmax's exponential with a multiply-add and an integer conversion, cutting vector work per score from 14 to 6 cycles. The post notes that FP8 makes matrix math 2x faster than BF16 but leaves softmax unchanged, making softmax the bottleneck in long-context prefill.

## Post

FP8 makes matrix math 2x faster than BF16 but does not speed up softmax, so softmax becomes the bottleneck in long-context prefill.

ZYFA's "lean P" replaces softmax's exponential with a multiply-add and an integer conversion, cutting vector work per score from 14 to 6 cycles.

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWaKN3gboAAeUKi.jpg&mode=thumb&exp=1794182400&sig=519c7642ec09d519)