Skip to content

Commit 20a560e

Browse files
How to enable experimental memory efficient attention on ROCm RDNA3.
1 parent 82c5308 commit 20a560e

File tree

1 file changed

+6
-0
lines changed

1 file changed

+6
-0
lines changed

README.md

+6
Original file line numberDiff line numberDiff line change
@@ -213,6 +213,12 @@ For 6700, 6600 and maybe other RDNA2 or older: ```HSA_OVERRIDE_GFX_VERSION=10.3.
213213

214214
For AMD 7600 and maybe other RDNA3 cards: ```HSA_OVERRIDE_GFX_VERSION=11.0.0 python main.py```
215215

216+
### AMD ROCm Tips
217+
218+
You can enable experimental memory efficient attention on pytorch 2.5 in ComfyUI on RDNA3 and potentially other AMD GPUs using this command:
219+
220+
```TORCH_ROCM_AOTRITON_ENABLE_EXPERIMENTAL=1 python main.py --use-pytorch-cross-attention```
221+
216222
# Notes
217223

218224
Only parts of the graph that have an output with all the correct inputs will be executed.

0 commit comments

Comments
 (0)