qlora_llama3_2_vision_11b

torchtune.models.llama3_2_vision.qlora_llama3_2_vision_11b(lora_attn_modules: List[Literal['q_proj', 'k_proj', 'v_proj', 'output_proj']], decoder_trainable: str = 'frozen', encoder_trainable: str = 'lora', fusion_trainable: str = 'lora', apply_lora_to_mlp: bool = False, apply_lora_to_output: bool = False, lora_rank: int = 8, lora_alpha: float = 16, lora_dropout: float = 0.0, use_dora: bool = False, *, quantize_base: bool = True, image_size: int = 560) → DeepFusionModel

用于创建启用QLoRA的Llama3.2 vision 11B模型的构建器。根据QLoRA论文，LoRA应用于线性层的基础模型权重被量化：https://arxiv.org/abs/2305.14314。请参阅lora_llama3_2_vision_11b以获取完整的API参数。