All-in-one node pack that collapses complex ComfyUI workflows into single nodes. Built for AI influencer content creation but useful for any image generation pipeline.

Nodes

| Node | What it replaces | Description | |------|-----------------|-------------| | Model Loader | UNETLoader + CLIPLoader + VAELoader + LoraLoader x2 | Pick a preset, get MODEL + CLIP + VAE. Supports LoRA stacking, sage attention, model sampling shift. | | Content Sampler | EmptyLatentImage + CLIPTextEncode x2 + KSampler + VAEDecode | All-in-one text-to-image. Prompt in, image out. | | Reference Sampler | VAEEncode + ReferenceLatent + CLIPTextEncode x2 + KSampler + VAEDecode | Reference-guided generation. Feed a face photo, get consistent identity across poses. | | Image Edit Sampler | TextEncodeQwenImageEditPlus + CLIPTextEncode + EmptyQwenImageLayeredLatent + KSampler + VAEDecode | Qwen image editing in one node. Feed reference images + edit prompt. | | Inpaint Sampler | VAEEncode + SetLatentNoiseMask + TextEncodeQwenImageEditPlus + CLIPTextEncode + KSampler + VAEDecode | Masked inpainting with Qwen Edit conditioning. Perfect for head swaps. | | Head Mask | LoadSAM3Model + SAM3Grounding + MaskGrow | SAM3-based head/object masking in one node. | | Upscaler | Multiple upscale nodes | All-in-one image upscaling. | | Face Swap | InsightFace + multiple detection/swap nodes | One-click face swap with InsightFace. | | Prompt Batch | CR Prompt List + CLIPTextEncode | Enter multiple prompts (one per line), get them all encoded for batch generation. | | Prompt Template | Manual prompt writing | AI influencer prompt generator with style presets. | | Dataset Saver | SaveImage + manual folder management | Save images to structured dataset folders with captions. | | Image Compare | Manual side-by-side comparison | Side-by-side image comparison output. |

Supported Model Presets

| Preset | Steps | Use Case | |--------|-------|----------| | Klein 9B | 6 | Fast reference-based generation | | Klein 9B True | 20 | High-quality reference generation | | Qwen Image Edit 2509 | 8 | Image editing, face swap, multi-angle | | Qwen Image Edit 2511 | 4* | Image editing (*with Lightning LoRA) | | Qwen Image 2512 | 8 | Text-to-image with shift 0.6 | | Z-Image Turbo | 6 | Fast text-to-image, detail pass | | Chroma HD | 3 | Fast high-quality generation |

Install

ComfyUI Manager (recommended)

Search for Fat Mex Nodes in ComfyUI Manager and click Install.

Manual

cd ComfyUI/custom_nodes
git clone https://github.com/FatMex/ComfyUI-FatMex-Nodes.git FatMex-Nodes
pip install -r FatMex-Nodes/requirements.txt

Dependencies

Some nodes require additional custom node packs:

Head Mask needs ComfyUI-SAM3
Face Swap needs InsightFace models (auto-downloaded on first use)

Restart ComfyUI after installation.

Workflows

Example workflows are included in the workflows/ folder. Load them directly in ComfyUI.

License

MIT License - see LICENSE