Llama-3.2-11B-Vision-Instruct-GGUF Support Request

[boricuapab]

Is this gguf llama 3.2 vision model supported?

https://huggingface.co/leafspark/Llama-3.2-11B-Vision-Instruct-GGUF/tree/main

I tried running the 20 gig 11b model from meta, but it's taking extremely long to run and never finishes on a 32 gig ram/16gig vram setup.

[heshengtao]

I found that the model you gave me doesn't seem to be able to be successfully called by llama cpp python, or it may be my version problem, I upgraded the VLM-GGUF loader node. Now you can choose the chat format, but if none of these options successfully load the clip model, I seriously doubt that llama cpp python does not support this model yet. https://github.com/abetlen/llama-cpp-python/tree/7ecdd944624cbd49e4af0a5ce1aa402607d58dcc?tab=readme-ov-file#multi-modal-models all the models written here can be called by party, because party relies on llama cpp python to call GGUF.

[power88]

Consider use Llama-3.2-vision-11b via Ollama? Now Ollama support Llama-3.2-vision-11b.