wangbo/ComfyUI
1
0
Fork 0
mirror of https://github.com/comfyanonymous/ComfyUI.git synced 2026-05-25 16:37:23 +08:00
Code Issues Actions 15 Packages Projects Releases Wiki Activity

Document --cuda-device comma format and MultiGPU Options relative_speed gap

Browse Source 
Two doc-only changes addressing minor CodeRabbit findings on PR #7063:

* cli_args.py: clarify --cuda-device help text to document the required comma-separated format ('0' or '0,1'), matching how the value is consumed by CUDA_VISIBLE_DEVICES in main.py.

* nodes_multigpu.py: add a docstring NOTE on the (currently unregistered) MultiGPUOptionsNode explaining that its relative_speed input is plumbed through to model_options['multigpu_options'] but is not yet consulted by the cond scheduler, which still uses uniform round-robin via next_available_device(). Wire relative_speed into the scheduler before re-enabling the node.

Amp-Thread-ID: https://ampcode.com/threads/T-019e43b8-8258-70fd-ab3a-53e4c97f85d5
Co-authored-by: Amp <amp@ampcode.com>
This commit is contained in:
Jedrzej Kosinski 2026-05-20 20:48:59 -07:00
parent ac0a90c323
commit 4d9106dced
2 changed files with 11 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
comfy/cli_args.py
View File

@ -49,7 +49,7 @@ parser.add_argument("--temp-directory", type=str, default=None, help="Set the Co
parser.add_argument("--input-directory", type=str, default=None, help="Set the ComfyUI input directory. Overrides --base-directory.") parser.add_argument("--input-directory", type=str, default=None, help="Set the ComfyUI input directory. Overrides --base-directory.")
parser.add_argument("--auto-launch", action="store_true", help="Automatically launch ComfyUI in the default browser.") parser.add_argument("--auto-launch", action="store_true", help="Automatically launch ComfyUI in the default browser.")
parser.add_argument("--disable-auto-launch", action="store_true", help="Disable auto launching the browser.") parser.add_argument("--disable-auto-launch", action="store_true", help="Disable auto launching the browser.")
parser.add_argument("--cuda-device", type=str, default=None, metavar="DEVICE_ID", help="Set the ids of cuda devices this instance will use. All other devices will not be visible.") parser.add_argument("--cuda-device", type=str, default=None, metavar="DEVICE_ID", help="Set the ids of cuda devices this instance will use, as a comma-separated list (e.g. '0' or '0,1'). All other devices will not be visible.")
parser.add_argument("--default-device", type=int, default=None, metavar="DEFAULT_DEVICE_ID", help="Set the id of the default device, all other devices will stay visible.") parser.add_argument("--default-device", type=int, default=None, metavar="DEFAULT_DEVICE_ID", help="Set the id of the default device, all other devices will stay visible.")
cm_group = parser.add_mutually_exclusive_group() cm_group = parser.add_mutually_exclusive_group()
cm_group.add_argument("--cuda-malloc", action="store_true", help="Enable cudaMallocAsync (enabled by default for torch 2.0 and up).") cm_group.add_argument("--cuda-malloc", action="store_true", help="Enable cudaMallocAsync (enabled by default for torch 2.0 and up).")

10
comfy_extras/nodes_multigpu.py
View File

@ -45,6 +45,16 @@ class MultiGPUCFGSplitNode(io.ComfyNode):
class MultiGPUOptionsNode(io.ComfyNode): class MultiGPUOptionsNode(io.ComfyNode):
""" """
Select the relative speed of GPUs in the special case they have significantly different performance from one another. Select the relative speed of GPUs in the special case they have significantly different performance from one another.
NOTE (not registered yet, see MultiGPUExtension.get_node_list below):
The output GPUOptionsGroup is plumbed through create_multigpu_deepclones() and stored on
model.model_options['multigpu_options'] via GPUOptionsGroup.register(), but the cond
scheduler in comfy/samplers.py (calc_cond_batch_outer_multigpu) does NOT yet consult
relative_speed when distributing conds across devices; it uses a uniform conds_per_device
round-robin via next_available_device(). Before re-enabling this node, wire its
relative_speed into the scheduler (e.g. via comfy.multigpu.load_balance_devices(),
which already implements the proportional split) so the input actually affects work
distribution.
""" """
@classmethod @classmethod