zhiqing
/

Qwen3-1.7B-INT8

Text Generation

text-generation-inference

8-bit precision

compressed-tensors

Model card Files Files and versions Community

Qwen3-1.7B-INT8

File size: 170 Bytes

9b267f2

default_stage:
  default_modifiers:
    SmoothQuantModifier: {smoothing_strength: 0.8}
    GPTQModifier:
      scheme: W8A8
      targets: Linear
      ignore: [lm_head]