unknown pre-tokenizer type: 'gpt-4o'
🥲 Failed to load the model
Failed to load model
error loading model: error loading model vocabulary: unknown pre-tokenizer type: 'gpt-4o'
Make sure you update to latest (needs beta runtime for lm studio)
Citing: https://lmstudio.ai/beta-releases... "No beta or experimental releases at this time. The stable version is the latest". I'm using LM Studio 0.3.10 (build 6) on Mac M1 (Apple Silicon).
By the way, the same issue I've been experiencing while trying to run this model via bare llama.cpp
llama_model_load: error loading model: error loading model vocabulary: unknown pre-tokenizer type: 'gpt-4o'
llama_model_load_from_file: failed to load model
common_init_from_params: failed to load model '/home/alex/ai/microsoft_Phi-4-mini-instruct-Q8_0.gguf'
srv load_model: failed to load model, '/home/alex/ai/microsoft_Phi-4-mini-instruct-Q8_0.gguf'
main: exiting due to model loading error
If you have some specific url for beta releases, I would appreciate if you provide it here.
Thank you.
it's not a beta release, it's a beta runtime
In LM Studio, enable "Power User" or "Developer" mode at the bottom left:
Then go to the "Developer" tab at the top left:
Then select "LM Runtimes" at the top:
Change to the "Beta" Runtime Extension Pack:
And then install the latest available runtime (should be v1.18.0)