GGUF
vllm
mistral-common
conversational

llama.cpp : error loading model architecture: unknown model architecture: 'mistral3'

#2
by alexaione - opened

Is anyone else facing an issue loading this model with llama.cpp?
The issue is seen on apple silicon running llama.cpp.
Previous versions of mistral ggufs are working fine, its happening only with the mistral-3 releases.
Below is the main error message we see...

print_info: file format = GGUF V3 (latest)
print_info: file type = Q8_0
print_info: file size = 8.40 GiB (8.50 BPW)
llama_model_load: error loading model: error loading model architecture: unknown model architecture: 'mistral3'
llama_model_load_from_file_impl: failed to load model

Fixed — issue caused by an older version of llama.cpp.

updated llama.cpp to Latest release: b7263

Note: Leaving this comment for anyone who encounters the same problem later.

alexaione changed discussion status to closed

Sign up or log in to comment