ComfyUI-GGUF Loader Nodes
Loads quantized GGUF diffusion and T5 weights directly into ComfyUI so 8 GB and 12 GB GPUs can run Flux and Hunyuan Video. Package: city96/ComfyUI-GGUF.
Updated Oct 17, 2026 · 1 min read
- Node / Pack
- ComfyUI-GGUF Loader Nodes
- Category
- custom-pack
- Inputs
- unet_name (.gguf), clip_name1, clip_name2
- Outputs
- MODEL, CLIP
- Source Pack
- city96/ComfyUI-GGUF
What ComfyUI-GGUF Loader Nodes does in a ComfyUI graph
Loads quantized GGUF diffusion and T5 weights directly into ComfyUI so 8 GB and 12 GB GPUs can run Flux and Hunyuan Video.
On the canvas, this node accepts `unet_name (.gguf), clip_name1, clip_name2` and emits `MODEL, CLIP`. Every pin is color-coded by data type so you cannot accidentally wire a pixel `IMAGE` tensor into a `LATENT` pin without an intermediate encode step.
Installation and missing-node recovery on Windows
ComfyUI-GGUF Loader Nodes is provided by `city96/ComfyUI-GGUF`. Open ComfyUI-Manager, click **Install Custom Nodes** (or **Install Missing Custom Nodes** if you loaded a workflow with red boxes), and restart the server.
Download the full Windows desktop or portable package at /download to get started with a clean Python 3.12 environment.
FAQs
How do I fix "ComfyUI-GGUF Loader Nodes" showing up as a red box in ComfyUI?
Open ComfyUI-Manager, click **Install Missing Custom Nodes**, install `city96/ComfyUI-GGUF`, and restart ComfyUI.
What are the main inputs and outputs of ComfyUI-GGUF Loader Nodes?
Inputs: unet_name (.gguf), clip_name1, clip_name2. Outputs: MODEL, CLIP.
Run on your hardware. Free forever.
The full ComfyUI engine: open source, fast, extensible, and yours to run however you want. Download the desktop app and open this workflow on your own machine.
