# \#mi350x

**URL:** https://devcommunity.amd.com/tag/mi350x/167.md

[Latest](https://devcommunity.amd.com/latest.md) · [Categories](https://devcommunity.amd.com/categories.md) · [Tags](https://devcommunity.amd.com/tags.md)

---

## [Efficiently Serving NVFP4 Models on AMD Instinct™ MI350X/MI355X Accelerators via Online NVFP4 to Quark MXFP4 Requantization](https://devcommunity.amd.com/t/efficiently-serving-nvfp4-models-on-amd-instinct-mi350x-mi355x-accelerators-via-online-nvfp4-to-quark-mxfp4-requantization/1049)

<div class="topic-metadata">

**Author:** [@nirth\_AMD](https://devcommunity.amd.com/u/nirth_AMD)\
**Replies:** 0\
**Last updated:** [September 18, 2026, 4:05pm UTC](https://devcommunity.amd.com/t/efficiently-serving-nvfp4-models-on-amd-instinct-mi350x-mi355x-accelerators-via-online-nvfp4-to-quark-mxfp4-requantization/1049 "2026-09-18T16:05:48Z")

</div>

What’s in it: A walkthrough of serving NVFP4-quantized models on AMD Instinct MI350X/MI355X by requantizing on the fly to MXFP4 via Quark inside SGLang - no separate preprocessing step, with minimal accuracy impact ve…