license: other license_name: qwen license_link: https://huggingface.co/huihui-ai/QVQ-72B-Preview-abliterated-GPTQ-Int8/blob/main/LICENSE language:

• en pipeline_tag: image-text-to-text base_model: huihui-ai/QVQ-72B-Preview-abliterated tags:
• abliterated
• uncensored
  • chat library_name: transformers

这是huihui-ai/QVQ-72B-Preview-abliterated的GPTQ量化8位版本。

本次仅为GPTQ量化测试，仅包含一个数据集："gptqmodel是一个基于GPTQ算法、具有友好API的易用模型量化库"。
如需自定义数据集，请联系我们：support@huihui.ai

使用说明

我们提供工具包助您便捷处理各类视觉输入（包括base64、URL及交错排列的图片视频），安装命令如下：

pip install qwen-vl-utils

以下代码片段展示如何结合transformers与qwen_vl_utils使用聊天模型：

from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
from qwen_vl_utils import process_vision_info

# 默认加载模型至可用设备
model = Qwen2VLForConditionalGeneration.from_pretrained(
    "huihui-ai/QVQ-72B-Preview-abliterated-GPTQ-Int8", torch_dtype="auto", device_map="auto"
)

# 默认处理器
processor = AutoProcessor.from_pretrained("huihui-ai/QVQ-72B-Preview-abliterated-GPTQ-Int8")

# 模型默认视觉token数范围为4-16384/图。可设置min_pixels与max_pixels调整范围（如256-1280个token）以平衡速度与显存
# min_pixels = 256*28*28
# max_pixels = 1280*28*28
# processor = AutoProcessor.from_pretrained("huihui-ai/QVQ-72B-Preview-abliterated-GPTQ-Int8", min_pixels=min_pixels, max_pixels=max_pixels)

messages = [
    {
        "role": "system",
        "content": [
            {"type": "text", "text": "你是一个无害且乐于助人的助手。你是阿里巴巴开发的Qwen模型，应当逐步思考问题。"}
        ],
    },
    {
        "role": "user",
        "content": [
            {
                "type": "image",
                "image": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/QVQ/demo.png",
            },
            {"type": "text", "text": "空白处应填入什么数值？"},
        ],
    }
]

# 推理准备
text = processor.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True
)
image_inputs, video_inputs = process_vision_info(messages)
inputs = processor(
    text=[text],
    images=image_inputs,
    videos=video_inputs,
    padding=True,
    return_tensors="pt",
)
inputs = inputs.to("cuda")

# 推理生成
generated_ids = model.generate(**inputs, max_new_tokens=8192)
generated_ids_trimmed = [
    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)
]
output_text = processor.batch_decode(
    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False
)
print(output_text)