unsloth/DeepSeek-V3-0324-GGUF-UD · Model loading error on Mac Studio (M3 Ultra)

Hi,
When deploying on my Mac Studio, I encountered the following error. Has anyone successfully run this model （Q5_K_M）?

[2025-05-08 10:55:00.900009] E llama_model_load: error loading model: check_tensor_dims: tensor 'blk.0.attn_q_b.weight' has wrong shape; expected  1536, 73728, got  1536, 24576,     1,     1
[2025-05-08 10:55:00.900009] E llama_model_load_from_file_impl: failed to load model
[2025-05-08 10:55:00.900013] E common_init_from_params: failed to load model '/usr/local/data/models/DeepSeek/unsloth/DeepSeek-V3-0324-GGUF-UD/Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00001-of-00010.gguf'

Device info:

2× Mac Studio with M3 Ultra, 512GB RAM
llama-server version: 5220 (cdf76586)
Built with: Apple clang version 17.0.0 (clang-1700.0.13.3) for arm64-apple-darwin24.4.0

Thanks!

unsloth
/

DeepSeek-V3-0324-GGUF-UD

Model loading error on Mac Studio (M3 Ultra) — tensor shape mismatch