Added example images for testing workflows:
- input_image.png (512x512) - for general upscaling workflows
- input_portrait.png (512x768) - for portrait/face upscaling workflows
Sound Lab's Musicgen_ node outputs AUDIO format that is only compatible with Sound Lab nodes like AudioPlay, not the built-in ComfyUI audio nodes (SaveAudio/PreviewAudio).
SaveAudio was erroring on 'waveform' key - the AUDIO output from
Musicgen_ node has a different internal structure than what SaveAudio
expects. PreviewAudio is more compatible with Sound Lab's AUDIO format.
Files are still saved to ComfyUI output directory, just through
PreviewAudio instead of SaveAudio.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Fixed medium, small, and melody workflows:
- Replaced non-existent nodes with Musicgen_ from Sound Lab
- Added missing links arrays to connect nodes properly
- Updated all metadata and performance specs
Note: Melody workflow simplified to text-only as Sound Lab doesn't
currently support melody conditioning via audio input.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Changed from non-existent nodes to actual Sound Lab nodes:
- Replaced MusicGenLoader/MusicGenTextEncode/MusicGenSampler with Musicgen_
- Replaced custom SaveAudio with standard SaveAudio node
- Added missing links array to connect nodes
- All parameters: prompt, duration, guidance_scale, seed, device
Node is called "Musicgen_" (with underscore) from comfyui-sound-lab.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
SD3.5 checkpoint doesn't contain CLIP encoders. Now using:
- CheckpointLoaderSimple for MODEL and VAE
- TripleCLIPLoader for CLIP-L, CLIP-G, and T5-XXL
- Standard CLIPTextEncode for prompts
This fixes the "clip input is invalid: None" error.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Replaced CheckpointLoaderSimple with UNETLoader + DualCLIPLoader.
Replaced CLIPTextEncode with CLIPTextEncodeFlux.
Added proper VAELoader with ae.safetensors.
Added ConditioningZeroOut for empty negative conditioning.
Removed old negative prompt input (FLUX doesn't use it).
Changes match FLUX Dev workflow structure.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Added FLUX VAE (ae.safetensors) to model configuration and updated
workflow to use it instead of non-existent pixel_space VAE.
This fixes the SaveImage data type error (1, 1, 16), |u1.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Node 3 (CLIPTextEncodeFlux) output feeds both KSampler (link 3) and
ConditioningZeroOut (link 8), so the output links array must include
both links.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
FLUX models require negative conditioning even though they don't use it.
Added ConditioningZeroOut node to create empty negative conditioning from
positive output, satisfying KSampler's required negative input.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Replace CheckpointLoaderSimple with UNETLoader
- Replace CLIPTextEncode with DualCLIPLoader + CLIPTextEncodeFlux
- Add VAELoader with pixel_space
- Remove negative prompt (FLUX uses guidance differently)
- Set CFG to 1.0, guidance in text encoder (3.5)
- Add all node connections in links array
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Both CLIP-L and T5-XXL come from the same repo, so they should
be in one entry with multiple files, not two separate entries.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Add CLIP-L and T5-XXL entries with type: clip so they get
linked to /workspace/ComfyUI/models/clip/ directory for
FLUX model usage via DualCLIPLoader.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
FLUX models need to be in /workspace/ComfyUI/models/unet/ directory
for UNETLoader to find them, not in diffusion_models directory.
Changed type for:
- black-forest-labs/FLUX.1-schnell
- black-forest-labs/FLUX.1-dev
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Replace DiffusersLoader with ImageOnlyCheckpointLoader
- Replace SVDSampler with SVD_img2vid_Conditioning + KSampler
- Add VideoLinearCFGGuidance for temporal consistency
- Add all node connections in links array
- Configure VHS_VideoCombine with correct parameters (25 frames)
- Increase steps to 30 for better quality with longer video
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Remove format-specific parameters from widgets_values array.
Only base parameters should be in widgets_values:
- frame_rate, loop_count, filename_prefix, format, pingpong, save_output
Format-specific params (pix_fmt, crf) are added dynamically by ComfyUI.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Replace DiffusersLoader with ImageOnlyCheckpointLoader
- Replace SVDSampler with SVD_img2vid_Conditioning + KSampler
- Add VideoLinearCFGGuidance for temporal consistency
- Add all node connections in links array
- Configure VHS_VideoCombine with H.264 parameters
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- CogVideoX-5b-I2V requires specific resolution (720x480)
- Cannot generate videos at different resolutions
- Update placeholder image to match model requirements
- Add enable_sequential_cpu_offload=true to DownloadAndLoadCogVideoModel
- Reduces VRAM from ~20GB to ~12GB at cost of slower inference
- Widget values: [model, precision, quantization, cpu_offload] = ['THUDM/CogVideoX-5b-I2V', 'bf16', 'disabled', true]
- Necessary for 24GB GPU with other services running
- Create 1024x1024 white placeholder with 'Input Frame' text
- Allows workflow validation without external image upload
- Will be replaced by API input in production use
- Add THUDM/CogVideoX-5b-I2V model entry (20GB, I2V-specific model)
- Add T5-XXL FP16 text encoder for CogVideoX (9GB)
- Add CLIP-L text encoder for CogVideoX and SD3 (1GB)
- Add CLIP-G text encoder for SD3 (3GB)
Note: CogVideoX models are auto-downloaded by DownloadAndLoadCogVideoModel node
Text encoders already linked manually to /workspace/ComfyUI/models/text_encoders/
- Change checkpoint from 'diffusers/stable-diffusion-xl-base-1.0' to 'sd_xl_base_1.0.safetensors'
- Change IPAdapter weight_type from 'original' to 'style transfer' (valid option)
- Fixes validation errors: invalid checkpoint name and invalid weight_type
- Change from 'source venv/bin/activate' to direct venv/bin/python execution
- Use exec to replace shell process with Python process
- Fixes issue where supervisor doesn't properly activate venv
- Ensures all extension dependencies are available
Added setup/comfyui-extensions-deps script that:
- Iterates through all custom_nodes directories
- Installs requirements.txt for each extension into ComfyUI venv
- Provides count of installed/skipped extensions
- Ensures all extension dependencies are met
This makes deployment reproducible and ensures all extensions
load without import errors.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- GitPython: for ComfyUI-Manager git operations
- opencv-python-headless: for image processing in extensions
- insightface: for face detection/recognition
- onnxruntime: for InsightFace models
- pyyaml: for config file parsing
- imageio-ffmpeg: for VideoHelperSuite
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Change CLIP vision model from ViT-H to ViT-bigG to match the
VIT-G preset in IPAdapterUnifiedLoader. This fixes dimension
mismatch error (1280 vs 768).
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
All services now use directory=. with relative paths in command,
matching the comfyui service configuration pattern.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Update supervisor configuration to use correct relative paths when
working directory is set to service subdirectory.
Changes:
- orchestrator: change to model-orchestrator directory, use venv/bin/python
- webdav-sync: already in webdav-sync directory, use venv/bin/python
This fixes "ENOENT" errors when supervisor tries to exec the python
interpreter from the virtual environment.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Update supervisord.conf to use venv Python executables for orchestrator and webdav-sync
- Add virtual environment setup to arty.yml setup/python-env script
- Isolate dependencies for webdav-sync, model-orchestrator, and vllm services
- Fixes dependency conflicts between services (pydantic versions)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Change CLIP H from h94/IP-Adapter to openai/clip-vit-large-patch14
- Change CLIP G from h94/IP-Adapter to laion/CLIP-ViT-bigG-14-laion2B-39B-b160k
- Update source paths to model.safetensors and open_clip_model.safetensors
- Fixes "header too large" error when loading CLIP vision models
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Changed CLIP Vision model source from openai/laion repos to h94/IP-Adapter
which provides the models in proper safetensors format that CLIPVisionLoader
can load directly.
Model sources:
- CLIP-ViT-H (SD 1.5): models/image_encoder/model.safetensors
- CLIP-ViT-bigG (SDXL): sdxl_models/image_encoder/model.safetensors
This fixes the "header too large" deserialization error caused by trying
to load PyTorch .bin files as safetensors.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
CLIPVisionLoader validates against a list of .safetensors files only.
Changed CLIP Vision model extensions from .bin back to .safetensors
in both the model configuration and workflow files.
Note: The actual files are PyTorch .bin format, but ComfyUI's
CLIPVisionLoader expects .safetensors naming convention. The symlinks
will point to the .bin files but use .safetensors extensions.
Changes:
- comfyui_models.yaml: Changed dest extensions to .safetensors
- Workflows: Changed clip_name to .safetensors extension
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
IPAdapterAdvanced requires a direct CLIPVision input, unlike the basic
IPAdapter node. Added CLIPVisionLoader nodes to both workflows:
Face workflow:
- Added CLIPVisionLoader (node 12) loading CLIP-ViT-bigG-14
- Connected to IPAdapterAdvanced (node 4) via link 20
Composition workflow:
- Added CLIPVisionLoader (node 15) loading CLIP-ViT-bigG-14
- Connected to both IPAdapterAdvanced nodes (6 and 7) via links 25 and 26
This provides the required CLIP Vision model for image understanding
in the IP-Adapter processing pipeline.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
Changed both IPAdapter nodes to IPAdapterAdvanced for better control
and consistency with the face workflow. IPAdapterAdvanced provides
additional parameters for fine-tuning the adapter behavior.
Updated widget values for both nodes:
- weight: 0.6/0.5 (composition influence strength)
- noise: 0.0
- weight_type: "ease in-out"
- combine_embeds: 1.0
- start_at: 0.0
- end_at: 1.0
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
