[pull] master from Comfy-Org:master

comfy/cli_args.py

@@ -145,6 +145,7 @@ def from_string(cls, value: str):
 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("--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("--disable-async-offload", action="store_true", help="Disable async weight offloading.")

comfy_api_nodes/nodes_sonilo.py

@@ -111,11 +111,10 @@ def define_schema(cls) -> IO.Schema:
                 ),
                 IO.Int.Input(
                     "duration",
-                    default=0,
-                    min=0,
+                    default=30,
+                    min=1,
                     max=360,
-                    tooltip="Target duration in seconds. Set to 0 to let the model "
-                    "infer the duration from the prompt. Maximum: 6 minutes.",
+                    tooltip="Target duration in seconds. Maximum: 6 minutes.",
                 ),
                 IO.Int.Input(
                     "seed",
@@ -150,14 +149,13 @@ def define_schema(cls) -> IO.Schema:
     async def execute(
         cls,
         prompt: str,
-        duration: int = 0,
+        duration: int = 1,
         seed: int = 0,
     ) -> IO.NodeOutput:
-        validate_string(prompt, strip_whitespace=True, min_length=1)
+        validate_string(prompt, strip_whitespace=True, min_length=1, max_length=1000)
         form = aiohttp.FormData()
         form.add_field("prompt", prompt)
-        if duration > 0:
-            form.add_field("duration", str(duration))
+        form.add_field("duration", str(duration))
         audio_bytes = await _stream_sonilo_music(
             cls,
             ApiEndpoint(path="/proxy/sonilo/t2m/generate", method="POST"),

main.py

@@ -55,7 +55,7 @@ def dump_traceback_on_sigint(signum, frame):
 import comfy_aimdo.control
 
 if enables_dynamic_vram():
-    comfy_aimdo.control.init()
+    comfy_aimdo.control.init(simple_vram_headroom=None if args.reserve_vram is None else int(args.reserve_vram * 1024 ** 3))
 
 if os.name == "nt":
     os.environ['MIMALLOC_PURGE_DELAY'] = '0'
@@ -231,7 +231,7 @@ def execute_script(script_path):
 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)):
         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':
             comfy_aimdo.control.set_log_debug()
         elif args.verbose == 'CRITICAL':

requirements.txt

@@ -1,5 +1,5 @@
 comfyui-frontend-package==1.45.15
-comfyui-workflow-templates==0.9.98
+comfyui-workflow-templates==0.10.0
 comfyui-embedded-docs==0.5.4
 torch
 torchsde
@@ -23,7 +23,7 @@ SQLAlchemy>=2.0.0
 filelock
 av>=16.0.0
 comfy-kitchen==0.2.10
-comfy-aimdo==0.4.9
+comfy-aimdo==0.4.10
 requests
 simpleeval>=1.0.0
 blake3