Files
QwQ-32B-Preview-quantized.w8a8/recipe.yaml
ModelHub XC 6c09572238 初始化项目,由ModelHub XC社区提供模型
Model: RedHatAI/QwQ-32B-Preview-quantized.w8a8
Source: Original Platform
2026-09-16 12:04:12 +08:00

8 lines
147 B
YAML

DEFAULT_stage:
DEFAULT_modifiers:
GPTQModifier:
targets: [Linear]
dampening_frac: 0.1
ignore: [lm_head]
scheme: W8A8