anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity

[ROCm][Perf] Add ROCm support for MXFP4 (w4a4) online quantization (#4104)

Browse Source 
Signed-off-by: Aleksi Vesanto <avesanto@amd.com>
Co-authored-by: WeiQing Chen <40507679+david6666666@users.noreply.github.com>
This commit is contained in:
Aleksi VesantoandWeiQing Chen authored and GitHub committed 2026-07-03 13:17:32 +08:00
1 parent 04f400d36c
commit 0455bffc37
4 files changed
+272 -3
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

docs/user_guide/quantization/mxfp4.md
+1 -1
View File
@@ -39,7 +39,7 @@ vLLM-Omni provides two quantization methods with different scale structures:
| NVIDIA Blackwell GPU (SM 100+) | ⭕ |
| NVIDIA Ada/Hopper GPU (SM 89+) | ⭕ |
| NVIDIA Ampere GPU (SM 80+) | ⭕ |
| AMD ROCm | ⭕ |
| AMD ROCm (gfx950 / MI355X) | ✅ |
| Intel XPU | ⭕ |
| Ascend NPU (Atlas 950 A5) | ✅ |