LM Studio: Failed to load model error
When I attempt to load the following model into LM Studio 0.3.5 (current version), the model OS-Atlas-Base-7B-GGUF/OS-Atlas-Base-7B.Q4_K_S.gguf from this site generates the following error: llama.cpp error: 'error loading model architecture: unknown model architecture: 'qwen2vl''
Any suggestions or expected updates?
When I attempt to load the following model into LM Studio 0.3.5 (current version), the model OS-Atlas-Base-7B-GGUF/OS-Atlas-Base-7B.Q4_K_S.gguf from this site generates the following error: llama.cpp error: 'error loading model architecture: unknown model architecture: 'qwen2vl''
Any suggestions or expected updates?
It works perfectly fine with latest llama.cpp - at least the next generation model quantized here without the vison model. First check if there is any development version available that comers with a more recent version of llama.cpp. If not either use llama.cpp directly or consider another user interface built on top of llama.cpp. If you need the vision capability make sure make sure to extract the vision model from the original model and separately provide it to llama.cpp or whatever UI you decide to use.