Does not run in Oobabooga?

#5
by AIGUYCONTENT - opened

I have 120GB of VRAM. Tried to load this model and got:

ValueError: Unknown quantization type, got exl2 - supported types are: ['awq', 'bitsandbytes_4bit', 'bitsandbytes_8bit', 'gptq', 'aqlm', 'quanto', 'eetq', 'hqq', 'fbgemm_fp8']

I know that oobabooga supports exl2. Is this an ooabooga error or?

You need to use the exllamav2 or exllamav2_HF loader

Sign up or log in to comment