Qwen3-8B-Instruct-2512-DFT-FP8/recipe.yaml

default_stage:
  default_modifiers:
    QuantizationModifier:
      targets: [Linear]
      ignore: [lm_head]
      scheme: FP8_DYNAMIC
初始化项目，由ModelHub XC社区提供模型 Model: JunHowie/Qwen3-8B-Instruct-2512-DFT-FP8 Source: Original Platform 2026-10-20 01:42:13 +08:00			`default_stage:`
			`default_modifiers:`
			`QuantizationModifier:`
			`targets: [Linear]`
			`ignore: [lm_head]`
			`scheme: FP8_DYNAMIC`