LM Studio: error loading model vocabulary
#2
by
janvalosek
- opened
Hey!
Thanks for the model! How can I use it in LM Studio? I'm getting a model vocabulary error when loading the model:
🥲 Failed to load the model
Failed to load model
error loading model: error loading model vocabulary: unknown pre-tokenizer type: 'csmpt'
error screenshot
Same problem here.
Happens for both BF16 and Q8_0 variants.
Tried with CUDA llama.cpp (Linux) v1.26.0 (beta) and CUDA llama.cpp (Linux) v1.25.2 (stable) runtimes in LM Studio.
Apologies, but this one will take more time, as we are currently overwhelmed with the end of the year :).
Wow, the end of the year, really? The current, or the previous one?
Sorry, just joking.:-) That was the first thing that came to my mind after reading your comment. But I understand your academic point of view.
I'm looking forward to try your model once the issue is resolved.
Thank you for your work guys! Keep doing.