Text Generation
Transformers
Safetensors
English
mixtral
conversational
text-generation-inference
4-bit precision
awq
Norobara-ZLoss-8x7B-AWQ / quant_config.json
TheBloke's picture
AWQ model commit
c12fef7
{
"zero_point": true,
"q_group_size": 128,
"w_bit": 4,
"version": "GEMM",
"modules_to_not_convert": [
"gate"
]
}