Merge branch 'master' into master

2025-11-30 14:25:19 +01:00
parent 3754d4e0a0 c829461cbf
commit 2d1769b9ef
6 changed files with 93 additions and 12 deletions
@@ -0,0 +1,28 @@
+name: Publish to Comfy registry
+on:
+  workflow_dispatch:
+  push:
+    branches:
+      - main
+      - master
+    paths:
+      - "pyproject.toml"
+
+permissions:
+  issues: write
+
+jobs:
+  publish-node:
+    name: Publish Custom Node to registry
+    runs-on: ubuntu-latest
+    if: ${{ github.repository_owner == 'erosDiffusion' }}
+    steps:
+      - name: Check out code
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+      - name: Publish Custom Node
+        uses: Comfy-Org/publish-node-action@v1
+        with:
+          ## Add your own personal access token to your Github Repository secrets and reference it here.
+          personal_access_token: ${{ secrets.REGISTRY_ACCESS_TOKEN }}
@@ -1 +1,21 @@
-empty...for now
+MIT License
+
+Copyright (c) 2025 erosDiffusion
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
@@ -1,27 +1,29 @@
 # FlowMatch Euler Discrete Scheduler for ComfyUI

-
 9 steps, big res, zero noise.

-**FlowMatchEulerDiscrete** seems not exposed in ComfyUI, but it is what the official demo in diffusers use.
+**FlowMatchEulerDiscrete** seems not exposed in ComfyUI, but it is what the official Z-Image demo in diffusers use.

 So:
+
 - I am exposing it in the scheduler section for you to use within KSampler.
 - On top I provide a node, experimental, to configure the scheduler for use with CustomSampler and play with.

 In short...if you want **sharper and noise free images**, use this!

 ## Installation
+
 - use comfy ui manager (search erosDiffusion or ComfyUI-EulerFlowMatchingDiscreteScheduler)
-  
+
 or

- ```git clone https://github.com/erosDiffusion/ComfyUI-EulerDiscreteScheduler.git``` in your custom nodes folder.
+- `git clone https://github.com/erosDiffusion/ComfyUI-EulerDiscreteScheduler.git` in your custom nodes folder.

 Example output (more below)
 <img width="1792" height="1120" alt="example" src="https://github.com/user-attachments/assets/91d4f679-9c9e-40ef-bed6-12bb860c37e7" />

 ## What you get
+
 - one new scheduler **FlowMatchEulerDiscreteScheduler** registered in the KSampler
 - a custom node that exposes all parameters of the FlowMatchEulerDiscreteScheduler which Outputs **SIGMAS** for use with **SamplerCustom** node.

@@ -36,17 +38,29 @@ Example output (more below)
  3. Adjust parameters to control the sampling behavior

 ## Tech bits:
+
 - https://huggingface.co/docs/diffusers/api/schedulers/flow_match_euler_discrete
 - https://huggingface.co/Tongyi-MAI/Z-Image-Turbo/blob/main/scheduler/scheduler_config.json

 ## Find this useful and want to support ?

-  [Buy me a beer!](https://donate.stripe.com/cNi9ALaASf65clXahPcV201)
-  
+[Buy me a beer!](https://donate.stripe.com/cNi9ALaASf65clXahPcV201)
+
 <img width="1920" height="1088" alt="ComfyUI_00716_" src="https://github.com/user-attachments/assets/bb2fc530-8a90-4180-96fb-adf6c48f5b09" />

 More examples:
 <img width="1536" height="1088" alt="image" src="https://github.com/user-attachments/assets/1ab561e7-b51d-413c-b788-13ed4fb6e129" />
 <img width="1536" height="1088" alt="image" src="https://github.com/user-attachments/assets/1931af7e-1b3e-47c9-ac20-27add5135a71" />

+## Changelog

+**1.0.3**
+
+- node publish action
+
+**1.0.2**
+
+- changed the device management in the custom scheduler node to be on gpu (cuda)
+- removed flash attention node dependency from the custom scheduler node
+- removed flash attention node from init
+- added mit licensing
@@ -133,6 +133,10 @@ class FlowMatchEulerSchedulerNode:
                    "default": "disable",
                    "tooltip": "Uses Karras noise schedule for smoother results. Similar to DPM++ samplers, often improves quality."
                }),
+                "device": (["auto", "cuda", "cpu"], {
+                    "default": "auto",
+                    "tooltip": "Device for sigma computation. 'auto' detects GPU if available, otherwise CPU. Using GPU avoids CPU->GPU transfers."
+                }),
            }
        }

@@ -161,6 +165,7 @@ class FlowMatchEulerSchedulerNode:
        use_dynamic_shifting,
        use_exponential_sigmas,
        use_karras_sigmas,
+        device="auto",
    ):
        # Convert string combo values to boolean
        config = {
@@ -184,6 +189,18 @@ class FlowMatchEulerSchedulerNode:
        
        # 1. Generate the full sigma schedule
        scheduler.set_timesteps(steps, device="cpu", mu=0.0)
+        # Determine device to use for sigma computation
+        if device == "auto":
+            # Auto-detect: use CUDA if available, otherwise CPU
+            target_device = "cuda" if torch.cuda.is_available() else "cpu"
+            print(f"[FlowMatch Scheduler] Auto-detected device: {target_device.upper()}")
+        else:
+            target_device = device
+            print(f"[FlowMatch Scheduler] Using manually specified device: {target_device.upper()}")
+        
+        # Set timesteps and get sigmas for the specified number of steps
+        # Using the model's device avoids unnecessary CPU->GPU transfers during sampling
+        scheduler.set_timesteps(steps, device=target_device, mu=0.0)
        sigmas = scheduler.sigmas
        
        # 2. Apply start_at_step and end_at_step (Slicing the sigmas tensor)
@@ -203,15 +220,17 @@ class FlowMatchEulerSchedulerNode:


 # Import Flash Attention node
-from .flash_attention_node import NODE_CLASS_MAPPINGS as FLASH_ATTN_MAPPINGS
-from .flash_attention_node import NODE_DISPLAY_NAME_MAPPINGS as FLASH_ATTN_DISPLAY_MAPPINGS
+# from .flash_attention_node import NODE_CLASS_MAPPINGS as FLASH_ATTN_MAPPINGS
+# from .flash_attention_node import NODE_DISPLAY_NAME_MAPPINGS as FLASH_ATTN_DISPLAY_MAPPINGS

 NODE_CLASS_MAPPINGS = {
    "FlowMatchEulerDiscreteScheduler (Custom)": FlowMatchEulerSchedulerNode,
-    **FLASH_ATTN_MAPPINGS
+    # **FLASH_ATTN_MAPPINGS
 }

 NODE_DISPLAY_NAME_MAPPINGS = {
    "FlowMatchEulerDiscreteScheduler (Custom)": "FlowMatch Euler Discrete Scheduler (Custom)",
    **FLASH_ATTN_DISPLAY_MAPPINGS
-}
+}
+    # **FLASH_ATTN_DISPLAY_MAPPINGS
+}
@@ -1,7 +1,7 @@
 [project]
 name = "erosdiffusion-eulerflowmatchingdiscretescheduler"
 description = "Noise Free images with Euler Discrete Scheduler in ComfyUI with Z-Image or other models"
-version = "1.0.1"
+version = "1.0.3"
 license = {file = "LICENSE.TXT"} 
 # classifiers = [
 #     # For OS-independent nodes (works on all operating systems)