Skip to content

Add env TORCH_AMD_CUDNN_ENABLED (third try) #10678

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
comfy-ovum wants to merge 3 commits into
base: master
Choose a base branch
from
+6 −2
Open

Add env TORCH_AMD_CUDNN_ENABLED (third try) #10678

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
417dbcb
Add env TORCH_AMD_CUDNN_ENABLED
sfinktah Nov 7, 2025
90ba86a
Add env TORCH_AMD_CUDNN_ENABLED (import os)
sfinktah Nov 8, 2025
de408d5
Add env TORCH_AMD_CUDNN_ENABLED (fix typo)
sfinktah Nov 8, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions comfy/model_management.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@
import platform
import weakref
import gc
import os

class VRAMState(Enum):
DISABLED = 0 #No vram present: no need to move models to vram
Expand Down Expand Up @@ -338,8 +339,11 @@ def amd_min_version(device=None, min_rdna_version=0):
if is_amd():
arch = torch.cuda.get_device_properties(get_torch_device()).gcnArchName
if not (any((a in arch) for a in AMD_RDNA2_AND_OLDER_ARCH)):
torch.backends.cudnn.enabled = False # Seems to improve things a lot on AMD
logging.info("Set: torch.backends.cudnn.enabled = False for better AMD performance.")
torch.backends.cudnn.enabled = os.environ.get("TORCH_AMD_CUDNN_ENABLED", "0").strip().lower() not in {
"0", "off", "false", "disable", "disabled", "no"}
if not torch.backends.cudnn.enabled:
logging.info(
"ComfyUI has set torch.backends.cudnn.enabled to False for better AMD performance. Set environment var TORCH_AMD_CUDNN_ENABLED=1 to enable it again.")

try:
rocm_version = tuple(map(int, str(torch.version.hip).split(".")[:2]))
Expand Down