Qwen2.5-VL-3B-Instruct-quantized.w8a8/recipe.yaml at main - Qwen2.5-VL-3B-Instruct-quantized.w8a8 - Gitea: Git with a cup of tea

RedHatAI/Qwen2.5-VL-3B-Instruct-quantized.w8a8

Files

ModelHub XC 08008235d8 初始化项目，由ModelHub XC社区提供模型

Model: RedHatAI/Qwen2.5-VL-3B-Instruct-quantized.w8a8
Source: Original Platform

2026-06-06 02:07:17 +08:00

8 lines

185 B

YAML

Raw Permalink Blame History

 DEFAULT_stage:
   DEFAULT_modifiers:
     GPTQModifier:
       sequential_targets: [Qwen2_5_VLDecoderLayer]
       scheme: W8A8
       targets: Linear
       ignore: [lm_head, 're:visual.*']