bartowski

/

Qwen_Qwen3.5-9B-GGUF

like 62

Yeah llama-cpp-python hasn't been updated in like half a year and Qwen3.5 is pretty new :(

Same for LM Studio, it doesn't support the new architecture of Qwen35.

lm studio's latest release does support Qwen35, can you double check you're on the latest @Abdelbasset ?

AttributeError: 'LlamaModel' object has no attribute 'sampler'

Any ideas?

use this.
https://github.com/JamePeng/llama-cpp-python

and re-install
pip install git+https://github.com/JamePeng/llama-cpp-python.git --no-cache-dir

and re-install
pip install git+https://github.com/JamePeng/llama-cpp-python.git --no-cache-dir

Do you perhaps know if this works with the GPU accelleration on Kaggle?
Or do you suggest another one?

Thanks

and re-install
pip install git+https://github.com/JamePeng/llama-cpp-python.git --no-cache-dir

Do you perhaps know if this works with the GPU accelleration on Kaggle?
Or do you suggest another one?

Thanks

Oh! Come to think of it, I did test it in a CPU-only environment.

No problem! I was working in CPU-only too.
But since Kaggle offers the accellerators, it would have been nice to use them.