MaziyarPanahi
/

Meta-Llama-3.1-70B-Instruct-GGUF

Text Generation

4-bit precision

8-bit precision

Model card Files Files and versions Community

Resources

View closed (4)

LLAma 3.1 8B f16 looks better then Q3_KM

#3 opened about 1 year ago by

Models Q4-K_M and Q5-K_M seems to use 'smaug-bpe' which fails to load the model.

#7 opened about 1 year ago by

IQ2_S model?

#4 opened about 1 year ago by