Files
SmolLM-135M-Instruct-quanti…/recipe.yaml
ModelHub XC 881c637af7 初始化项目,由ModelHub XC社区提供模型
Model: RedHatAI/SmolLM-135M-Instruct-quantized.w8a16
Source: Original Platform
2026-10-04 08:35:13 +08:00

9 lines
174 B
YAML

quant_stage:
quant_modifiers:
GPTQModifier:
sequential_update: false
dampening_frac: 0.01
ignore: [lm_head]
scheme: W8A16
targets: Linear