ValueError: Received 126 parameters not in model.
#1
by p215636 - opened
I’m trying to run the model NexVeridian/gemma-4-E4B-8bit as the model card suggests and I’m getting a ValueError: Received 126 parameters not in model
My MLX version is 0.31.2 and MLX-LM is 0.31.3
I resort to pip install --upgrade --no-cache-dir mlx-lm mlx in the hope that this would clean the slate and resolve any conflicts, but regrettable no success. Can you kindly confirm if it still works on your setup or provide some suggestions as to what I’m missing…
Thanking you in advance…