Quantized with these parameters:

--bits 4

--group_size 128

--desc_act 1

--damp 0.1

--seqlen 16384

--num_samples 512

Quantization Dataset: Erotiquant XL

Downloads last month
21
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Space using openerotica/Llama-3-lima-nsfw-16k-test-GPTQ 1