wangbo/ComfyUI
1
0
Fork 0
mirror of https://github.com/comfyanonymous/ComfyUI.git synced 2026-06-24 00:39:30 +08:00
Code Issues Actions 18 Packages Projects Releases Wiki Activity

main: implement --vram-headroom

Browse Source 
Implement --vram-headroom for dynamic vram as a hybrid debug/diagnostic
option that can be used for people who still report shared VRAM spills.
They can trial and error the setting to maintain a bit more headroom to
avoid shared VRAM spills.
This commit is contained in:
Rattus 2026-06-15 19:43:15 +10:00
parent b664349ae7
commit 002f2bedd6
3 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
comfy/cli_args.py
View File

@ -145,6 +145,7 @@ vram_group.add_argument("--novram", action="store_true", help="When lowvram isn'
vram_group.add_argument("--cpu", action="store_true", help="To use the CPU for everything (slow).") vram_group.add_argument("--cpu", action="store_true", help="To use the CPU for everything (slow).")
parser.add_argument("--reserve-vram", type=float, default=None, help="Set the amount of vram in GB you want to reserve for use by your OS/other software. By default some amount is reserved depending on your OS.") parser.add_argument("--reserve-vram", type=float, default=None, help="Set the amount of vram in GB you want to reserve for use by your OS/other software. By default some amount is reserved depending on your OS.")
parser.add_argument("--vram-headroom", type=float, default=0, help="Set the amount of vram in GB for DynamicVRAM to maintain as extra headroom above default. ComfyUI will try and keep this much VRAM completely free and unused, even counting VRAM from other apps.")
parser.add_argument("--async-offload", nargs='?', const=2, type=int, default=None, metavar="NUM_STREAMS", help="Use async weight offloading. An optional argument controls the amount of offload streams. Default is 2. Enabled by default on Nvidia.") parser.add_argument("--async-offload", nargs='?', const=2, type=int, default=None, metavar="NUM_STREAMS", help="Use async weight offloading. An optional argument controls the amount of offload streams. Default is 2. Enabled by default on Nvidia.")
parser.add_argument("--disable-async-offload", action="store_true", help="Disable async weight offloading.") parser.add_argument("--disable-async-offload", action="store_true", help="Disable async weight offloading.")

2
main.py
View File

@ -231,7 +231,7 @@ import comfy.model_patcher
if args.enable_dynamic_vram or (enables_dynamic_vram() and comfy.model_management.is_nvidia() and not comfy.model_management.is_wsl()): if args.enable_dynamic_vram or (enables_dynamic_vram() and comfy.model_management.is_nvidia() and not comfy.model_management.is_wsl()):
if (not args.enable_dynamic_vram) and (comfy.model_management.torch_version_numeric < (2, 8)): if (not args.enable_dynamic_vram) and (comfy.model_management.torch_version_numeric < (2, 8)):
logging.warning("Unsupported Pytorch detected. DynamicVRAM support requires Pytorch version 2.8 or later. Falling back to legacy ModelPatcher. VRAM estimates may be unreliable especially on Windows") logging.warning("Unsupported Pytorch detected. DynamicVRAM support requires Pytorch version 2.8 or later. Falling back to legacy ModelPatcher. VRAM estimates may be unreliable especially on Windows")
elif comfy_aimdo.control.init_devices(d.index for d in comfy.model_management.get_all_torch_devices()): elif comfy_aimdo.control.init_devices((d.index, int(args.vram_headroom * 1024 ** 3)) for d in comfy.model_management.get_all_torch_devices()):
if args.verbose == 'DEBUG': if args.verbose == 'DEBUG':
comfy_aimdo.control.set_log_debug() comfy_aimdo.control.set_log_debug()
elif args.verbose == 'CRITICAL': elif args.verbose == 'CRITICAL':

2
requirements.txt
View File

@ -23,7 +23,7 @@ SQLAlchemy>=2.0.0
filelock filelock
av>=16.0.0 av>=16.0.0
comfy-kitchen==0.2.10 comfy-kitchen==0.2.10
comfy-aimdo==0.4.9 comfy-aimdo==0.4.10
requests requests
simpleeval>=1.0.0 simpleeval>=1.0.0
blake3 blake3