ComfyUI Node
Vision Settings (GGUF)
A ComfyUI node in Kinburg-Nodes/LLM with 4 inputs and 1 output.
Vision Settings (GGUF)
- vision
◄mmproj▾►
◄mmproj_path►
◄vision_handlerauto (MTMD)►
◄image_max_side1024►
CategoryKinburg-Nodes/LLM
Inputs (4)
| Name | Type | Default | Description |
|---|---|---|---|
| mmproj | COMBO | Projector mmproj .gguf from ComfyUI/models/llm (subfolders included, so you can keep a model + its mmproj together in one folder; mmproj-named files first). Choose the placeholder to type a path in mmproj_path. | |
| mmproj_path | STRING | Full path to the mmproj .gguf (when mmproj is the placeholder). Surrounding quotes are stripped. | |
| vision_handler | COMBO | auto (MTMD) | auto (MTMD) is llama.cpp's generic multimodal loader and fits most modern vision GGUFs. Switch to the model's family only if auto fails. |
| image_max_side | INT | 10240–4096 | Downscale each image so its longest side is at most this many px before sending. 0 = full size. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| vision | KINBURG_VISION_CONFIG | — |