Merge c6ad7a4d01 into 22ad513c72

* add dependency aware cache that removed a cached node as soon as all of its decendents have executed. This allows users with lower RAM to run workflows they would otherwise not be able to run. The downside is that every workflow will fully run each time even if no nodes have changed.

* remove test code

* tidy code

.github/workflows/stable-release.yml

@@ -1,4 +1,3 @@
-
 name: "Release Stable Version"
 
 on:

.github/workflows/test-unit.yml

@@ -15,7 +15,7 @@ jobs:
     continue-on-error: true
     steps:
     - uses: actions/checkout@v4
-    - name: Set up Python      
+    - name: Set up Python
       uses: actions/setup-python@v4
       with:
         python-version: '3.12'

README.md

@@ -36,7 +36,7 @@ ComfyUI lets you design and execute advanced stable diffusion pipelines using a
 ## Get Started
 
 #### [Desktop Application](https://www.comfy.org/download)
-- The easiest way to get started. 
+- The easiest way to get started.
 - Available on Windows & macOS.
 
 #### [Windows Portable Package](#installing)
@@ -190,7 +190,7 @@ This is the command to install the nightly with ROCm 6.3 which might have some p
 ### Intel GPUs (Windows and Linux)
 
 (Option 1) Intel Arc GPU users can install native PyTorch with torch.xpu support using pip (currently available in PyTorch nightly builds). More information can be found [here](https://pytorch.org/docs/main/notes/get_start_xpu.html)
-  
+
 1. To install PyTorch nightly, use the following command:
 
 ```pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/xpu```
@@ -321,7 +321,7 @@ Generate a self-signed certificate (not appropriate for shared/production use) a
 
 Use `--tls-keyfile key.pem --tls-certfile cert.pem` to enable TLS/SSL, the app will now be accessible with `https://...` instead of `http://...`.
 
-> Note: Windows users can use [alexisrolland/docker-openssl](https://github.com/alexisrolland/docker-openssl) or one of the [3rd party binary distributions](https://wiki.openssl.org/index.php/Binaries) to run the command example above. 
+> Note: Windows users can use [alexisrolland/docker-openssl](https://github.com/alexisrolland/docker-openssl) or one of the [3rd party binary distributions](https://wiki.openssl.org/index.php/Binaries) to run the command example above.
 <br/><br/>If you use a container, note that the volume mount `-v` can be a relative path so `... -v ".\:/openssl-certs" ...` would create the key & cert files in the current directory of your command prompt or powershell terminal.
 
 ## Support and dev channel

comfy/cli_args.py

@@ -101,6 +101,7 @@ parser.add_argument("--preview-size", type=int, default=512, help="Sets the maxi
 cache_group = parser.add_mutually_exclusive_group()
 cache_group.add_argument("--cache-classic", action="store_true", help="Use the old style (aggressive) caching.")
 cache_group.add_argument("--cache-lru", type=int, default=0, help="Use LRU caching with a maximum of N node results cached. May use more RAM/VRAM.")
+cache_group.add_argument("--cache-none", action="store_true", help="Reduced RAM/VRAM usage at the expense of executing every node for each run.")
 
 attn_group = parser.add_mutually_exclusive_group()
 attn_group.add_argument("--use-split-cross-attention", action="store_true", help="Use the split cross attention optimization. Ignored when xformers is used.")

comfy/hooks.py

@@ -587,7 +587,7 @@ def get_sorted_list_via_attr(objects: list, attr: str) -> list:
         sorted_list.extend(object_list)
     return sorted_list
 
-def create_transformer_options_from_hooks(model: ModelPatcher, hooks: HookGroup,  transformer_options: dict[str]=None):
+def create_transformer_options_from_hooks(model: ModelPatcher, hooks: HookGroup, transformer_options: dict[str]=None):
     # if no hooks or is not a ModelPatcher for sampling, return empty dict
     if hooks is None or model.is_clip:
         return {}

comfy/supported_models.py

@@ -618,10 +618,10 @@ class PixArtAlpha(supported_models_base.BASE):
     }
 
     sampling_settings = {
-        "beta_schedule" : "sqrt_linear",
-        "linear_start"  : 0.0001,
-        "linear_end"    : 0.02,
-        "timesteps"     : 1000,
+        "beta_schedule": "sqrt_linear",
+        "linear_start": 0.0001,
+        "linear_end": 0.02,
+        "timesteps": 1000,
     }
 
     unet_extra_config = {}
@@ -681,8 +681,8 @@ class HunyuanDiT1(HunyuanDiT):
     unet_extra_config = {}
 
     sampling_settings = {
-        "linear_start" : 0.00085,
-        "linear_end" : 0.03,
+        "linear_start": 0.00085,
+        "linear_end": 0.03,
     }
 
 class Flux(supported_models_base.BASE):

comfy/utils.py

@@ -449,8 +449,8 @@ PIXART_MAP_BLOCK = {
     ("mlp.fc1.bias", "ff.net.0.proj.bias"),
     ("mlp.fc2.weight", "ff.net.2.weight"),
     ("mlp.fc2.bias", "ff.net.2.bias"),
-    ("cross_attn.proj.weight" ,"attn2.to_out.0.weight"),
-    ("cross_attn.proj.bias"   ,"attn2.to_out.0.bias"),
+    ("cross_attn.proj.weight", "attn2.to_out.0.weight"),
+    ("cross_attn.proj.bias", "attn2.to_out.0.bias"),
 }
 
 def pixart_to_diffusers(mmdit_config, output_prefix=""):

comfy_execution/caching.py

@@ -316,3 +316,156 @@ class LRUCache(BasicCache):
             self.children[cache_key].append(self.cache_key_set.get_data_key(child_id))
         return self
 
+
+class DependencyAwareCache(BasicCache):
+    """
+    A cache implementation that tracks dependencies between nodes and manages
+    their execution and caching accordingly. It extends the BasicCache class.
+    Nodes are removed from this cache once all of their descendants have been
+    executed.
+    """
+
+    def __init__(self, key_class):
+        """
+        Initialize the DependencyAwareCache.
+
+        Args:
+            key_class: The class used for generating cache keys.
+        """
+        super().__init__(key_class)
+        self.descendants = {}  # Maps node_id -> set of descendant node_ids
+        self.ancestors = {}    # Maps node_id -> set of ancestor node_ids
+        self.executed_nodes = set()  # Tracks nodes that have been executed
+
+    def set_prompt(self, dynprompt, node_ids, is_changed_cache):
+        """
+        Clear the entire cache and rebuild the dependency graph.
+
+        Args:
+            dynprompt: The dynamic prompt object containing node information.
+            node_ids: List of node IDs to initialize the cache for.
+            is_changed_cache: Flag indicating if the cache has changed.
+        """
+        # Clear all existing cache data
+        self.cache.clear()
+        self.subcaches.clear()
+        self.descendants.clear()
+        self.ancestors.clear()
+        self.executed_nodes.clear()
+
+        # Call the parent method to initialize the cache with the new prompt
+        super().set_prompt(dynprompt, node_ids, is_changed_cache)
+
+        # Rebuild the dependency graph
+        self._build_dependency_graph(dynprompt, node_ids)
+
+    def _build_dependency_graph(self, dynprompt, node_ids):
+        """
+        Build the dependency graph for all nodes.
+
+        Args:
+            dynprompt: The dynamic prompt object containing node information.
+            node_ids: List of node IDs to build the graph for.
+        """
+        self.descendants.clear()
+        self.ancestors.clear()
+        for node_id in node_ids:
+            self.descendants[node_id] = set()
+            self.ancestors[node_id] = set()
+
+        for node_id in node_ids:
+            inputs = dynprompt.get_node(node_id)["inputs"]
+            for input_data in inputs.values():
+                if is_link(input_data):  # Check if the input is a link to another node
+                    ancestor_id = input_data[0]
+                    self.descendants[ancestor_id].add(node_id)
+                    self.ancestors[node_id].add(ancestor_id)
+
+    def set(self, node_id, value):
+        """
+        Mark a node as executed and store its value in the cache.
+
+        Args:
+            node_id: The ID of the node to store.
+            value: The value to store for the node.
+        """
+        self._set_immediate(node_id, value)
+        self.executed_nodes.add(node_id)
+        self._cleanup_ancestors(node_id)
+
+    def get(self, node_id):
+        """
+        Retrieve the cached value for a node.
+
+        Args:
+            node_id: The ID of the node to retrieve.
+
+        Returns:
+            The cached value for the node.
+        """
+        return self._get_immediate(node_id)
+
+    def ensure_subcache_for(self, node_id, children_ids):
+        """
+        Ensure a subcache exists for a node and update dependencies.
+
+        Args:
+            node_id: The ID of the parent node.
+            children_ids: List of child node IDs to associate with the parent node.
+
+        Returns:
+            The subcache object for the node.
+        """
+        subcache = super()._ensure_subcache(node_id, children_ids)
+        for child_id in children_ids:
+            self.descendants[node_id].add(child_id)
+            self.ancestors[child_id].add(node_id)
+        return subcache
+
+    def _cleanup_ancestors(self, node_id):
+        """
+        Check if ancestors of a node can be removed from the cache.
+
+        Args:
+            node_id: The ID of the node whose ancestors are to be checked.
+        """
+        for ancestor_id in self.ancestors.get(node_id, []):
+            if ancestor_id in self.executed_nodes:
+                # Remove ancestor if all its descendants have been executed
+                if all(descendant in self.executed_nodes for descendant in self.descendants[ancestor_id]):
+                    self._remove_node(ancestor_id)
+
+    def _remove_node(self, node_id):
+        """
+        Remove a node from the cache.
+
+        Args:
+            node_id: The ID of the node to remove.
+        """
+        cache_key = self.cache_key_set.get_data_key(node_id)
+        if cache_key in self.cache:
+            del self.cache[cache_key]
+        subcache_key = self.cache_key_set.get_subcache_key(node_id)
+        if subcache_key in self.subcaches:
+            del self.subcaches[subcache_key]
+
+    def clean_unused(self):
+        """
+        Clean up unused nodes. This is a no-op for this cache implementation.
+        """
+        pass
+
+    def recursive_debug_dump(self):
+        """
+        Dump the cache and dependency graph for debugging.
+
+        Returns:
+            A list containing the cache state and dependency graph.
+        """
+        result = super().recursive_debug_dump()
+        result.append({
+            "descendants": self.descendants,
+            "ancestors": self.ancestors,
+            "executed_nodes": list(self.executed_nodes),
+        })
+        return result

comfy_extras/nodes_morphology.py

@@ -9,7 +9,7 @@ class Morphology:
     @classmethod
     def INPUT_TYPES(s):
         return {"required": {"image": ("IMAGE",),
-                                "operation": (["erode",  "dilate", "open", "close", "gradient", "bottom_hat", "top_hat"],),
+                                "operation": (["erode", "dilate", "open", "close", "gradient", "bottom_hat", "top_hat"],),
                                 "kernel_size": ("INT", {"default": 3, "min": 3, "max": 999, "step": 1}),
                                 }}
 

execution.py

@@ -15,7 +15,7 @@ import nodes
 import comfy.model_management
 from comfy_execution.graph import get_input_info, ExecutionList, DynamicPrompt, ExecutionBlocker
 from comfy_execution.graph_utils import is_link, GraphBuilder
-from comfy_execution.caching import HierarchicalCache, LRUCache, CacheKeySetInputSignature, CacheKeySetID
+from comfy_execution.caching import HierarchicalCache, LRUCache, DependencyAwareCache, CacheKeySetInputSignature, CacheKeySetID
 from comfy_execution.validation import validate_node_input
 
 class ExecutionResult(Enum):
@@ -59,20 +59,27 @@ class IsChangedCache:
             self.is_changed[node_id] = node["is_changed"]
         return self.is_changed[node_id]
 
-class CacheSet:
-    def __init__(self, lru_size=None):
-        if lru_size is None or lru_size == 0:
-            self.init_classic_cache()
-        else:
-            self.init_lru_cache(lru_size)
-        self.all = [self.outputs, self.ui, self.objects]
 
-    # Useful for those with ample RAM/VRAM -- allows experimenting without
-    # blowing away the cache every time
-    def init_lru_cache(self, cache_size):
-        self.outputs = LRUCache(CacheKeySetInputSignature, max_size=cache_size)
-        self.ui = LRUCache(CacheKeySetInputSignature, max_size=cache_size)
-        self.objects = HierarchicalCache(CacheKeySetID)
+class CacheType(Enum):
+    CLASSIC = 0
+    LRU = 1
+    DEPENDENCY_AWARE = 2
+
+
+class CacheSet:
+    def __init__(self, cache_type=None, cache_size=None):
+        if cache_type == CacheType.DEPENDENCY_AWARE:
+            self.init_dependency_aware_cache()
+            logging.info("Disabling intermediate node cache.")
+        elif cache_type == CacheType.LRU:
+            if cache_size is None:
+                cache_size = 0
+            self.init_lru_cache(cache_size)
+            logging.info("Using LRU cache")
+        else:
+            self.init_classic_cache()
+
+        self.all = [self.outputs, self.ui, self.objects]
 
     # Performs like the old cache -- dump data ASAP
     def init_classic_cache(self):
@@ -80,6 +87,17 @@ class CacheSet:
         self.ui = HierarchicalCache(CacheKeySetInputSignature)
         self.objects = HierarchicalCache(CacheKeySetID)
 
+    def init_lru_cache(self, cache_size):
+        self.outputs = LRUCache(CacheKeySetInputSignature, max_size=cache_size)
+        self.ui = LRUCache(CacheKeySetInputSignature, max_size=cache_size)
+        self.objects = HierarchicalCache(CacheKeySetID)
+
+    # only hold cached items while the decendents have not executed
+    def init_dependency_aware_cache(self):
+        self.outputs = DependencyAwareCache(CacheKeySetInputSignature)
+        self.ui = DependencyAwareCache(CacheKeySetInputSignature)
+        self.objects = DependencyAwareCache(CacheKeySetID)
+
     def recursive_debug_dump(self):
         result = {
             "outputs": self.outputs.recursive_debug_dump(),
@@ -414,13 +432,14 @@ def execute(server, dynprompt, caches, current_item, extra_data, executed, promp
     return (ExecutionResult.SUCCESS, None, None)
 
 class PromptExecutor:
-    def __init__(self, server, lru_size=None):
-        self.lru_size = lru_size
+    def __init__(self, server, cache_type=False, cache_size=None):
+        self.cache_size = cache_size
+        self.cache_type = cache_type
         self.server = server
         self.reset()
 
     def reset(self):
-        self.caches = CacheSet(self.lru_size)
+        self.caches = CacheSet(cache_type=self.cache_type, cache_size=self.cache_size)
         self.status_messages = []
         self.success = True
 

folder_paths.py

@@ -84,7 +84,7 @@ class CacheHelper:
 cache_helper = CacheHelper()
 
 extension_mimetypes_cache = {
-    "webp" : "image",
+    "webp": "image",
 }
 
 def map_legacy(folder_name: str) -> str:

main.py

@@ -156,7 +156,13 @@ def cuda_malloc_warning():
 
 def prompt_worker(q, server_instance):
     current_time: float = 0.0
-    e = execution.PromptExecutor(server_instance, lru_size=args.cache_lru)
+    cache_type = execution.CacheType.CLASSIC
+    if args.cache_lru > 0:
+        cache_type = execution.CacheType.LRU
+    elif args.cache_none:
+        cache_type = execution.CacheType.DEPENDENCY_AWARE
+
+    e = execution.PromptExecutor(server_instance, cache_type=cache_type, cache_size=args.cache_lru)
     last_gc_collect = 0
     need_gc = False
     gc_collect_interval = 10.0

nodes.py

@@ -1989,7 +1989,7 @@ NODE_CLASS_MAPPINGS = {
     "ImageBatch": ImageBatch,
     "ImagePadForOutpaint": ImagePadForOutpaint,
     "EmptyImage": EmptyImage,
-    "ConditioningAverage": ConditioningAverage ,
+    "ConditioningAverage": ConditioningAverage,
     "ConditioningCombine": ConditioningCombine,
     "ConditioningConcat": ConditioningConcat,
     "ConditioningSetArea": ConditioningSetArea,
@@ -2076,7 +2076,7 @@ NODE_DISPLAY_NAME_MAPPINGS = {
     "LatentUpscaleBy": "Upscale Latent By",
     "LatentComposite": "Latent Composite",
     "LatentBlend": "Latent Blend",
-    "LatentFromBatch" : "Latent From Batch",
+    "LatentFromBatch": "Latent From Batch",
     "RepeatLatentBatch": "Repeat Latent Batch",
     # Image
     "SaveImage": "Save Image",

pytest.ini

@@ -1,5 +1,5 @@
 [pytest]
-markers = 
+markers =
   inference: mark as inference test (deselect with '-m "not inference"')
   execution: mark as execution test (deselect with '-m "not execution"')
 testpaths =

requirements.txt

@@ -1,4 +1,4 @@
-comfyui-frontend-package==1.14.6
+comfyui-frontend-package==1.15.13
 torch
 torchsde
 torchvision

server.py

@@ -48,7 +48,7 @@ async def send_socket_catch_exception(function, message):
 @web.middleware
 async def cache_control(request: web.Request, handler):
     response: web.Response = await handler(request)
-    if request.path.endswith('.js') or request.path.endswith('.css'):
+    if request.path.endswith('.js') or request.path.endswith('.css') or request.path.endswith('index.json'):
         response.headers.setdefault('Cache-Control', 'no-cache')
     return response