error loading in LM Studio

#1
by Conrad3000 - opened

When I try to load in LM Studio I get the following error: llama.cpp error: 'error loading model vocabulary: unknown pre-tokenizer type: 'deepseek-r1-qwen''

updating my llama.cpp runtime (cuda in my case) seems to have fixed the issue

I am getting the same error, what does ". I'm on LM Studio 0.3.7; which is the latest release.

Note I am using the model "DeepSeek-R1-Distill-Qwen-7B-GGUF/DeepSeek-R1-Distill-Qwen-7B-Q4_K_M.gguf" from lmstudio-community.

How do you update the llama.cpp runtime?

Ok - it is now loading. Thsi is what O did:

  1. Click to the gear icon in the bottom RIGHT corner of the chat.

image.png

  1. Clicked on "Runtimes":

image.png

  1. Updated the runtimes:

image.png

I am getting the same error, what does ". I'm on LM Studio 0.3.7; which is the latest release.

Note I am using the model "DeepSeek-R1-Distill-Qwen-7B-GGUF/DeepSeek-R1-Distill-Qwen-7B-Q4_K_M.gguf" from lmstudio-community.

How do you update the llama.cpp runtime?

Ok - it is now loading. Thsi is what O did:

  1. Click to the gear icon in the bottom RIGHT corner of the chat.

image.png

  1. Clicked on "Runtimes":

image.png

  1. Updated the runtimes:

image.png

you Sir, are a champ! Fixed it for me

Sign up or log in to comment