[Recipe] Document MiniMax H3 ROCm (gfx950) BF16 serving (#5723)

Signed-off-by: Xiao Yu <xiao.yu@amd.com>
Signed-off-by: amd-xiaoyu12 <xiao.yu@amd.com>
This commit is contained in:
amd-xiaoyu12 authored and GitHub committed 2026-08-09 22:18:24 +08:00
1 parent 592637015f
commit f01e4bdddc
2 files changed
+121 -5

No files matched your search

+7
View File
@@ -97,3 +97,10 @@ th {
|`HiDreamImagePipeline` | HiDream-I1-Full | `HiDream-ai/HiDream-I1-Full` | ✅︎ | ✅︎ | | |
✅︎ indicates the model is supported on that backend. Empty cells mean not listed as supported on that backend.
<sup>H3</sup> MiniMax H3 AMD GPU support is validated on gfx942 (MI300X) and
gfx950 (MI350) in BF16, using the AITER `FLASH_ATTN` diffusion attention backend.
Other AMD SKUs (e.g. MI325X) are not listed until their
own validation evidence is added. See the
[MiniMax H3 recipe](https://recipes.vllm.ai/MiniMaxAI/MiniMax-H3) for
per-architecture serving commands.