stan-hua
/

Mistral-7B-Instruct-v0.3-AWQ-W4A16

4-bit precision

Model card Files Files and versions Community

Mistral-7B-Instruct-v0.3-AWQ-W4A16 / quant_config.json

stan-hua's picture

AWQ model for mistralai/Mistral-7B-Instruct-v0.3: {'w_bit': 4, 'zero_point': True, 'q_group_size': 128, 'version': 'GEMM'}

82b4498 verified 7 days ago

history blame contribute delete

90 Bytes

	{
	"w_bit": 4,
	"zero_point": true,
	"q_group_size": 128,
	"version": "GEMM"
	}