The original model is here: https://huggingface.co./Lin-Chen/ShareGPT4V-7B
This is the K-type quantized variant (for inference with llama.cpp llava-cli)

In my tests this is currently the best llava based vision model

Downloads last month
22
GGUF
Model size
312M params
Architecture
clip

6-bit

16-bit

Inference API
Unable to determine this model's library. Check the docs .