语言:

• 英文 库名称: transformers 许可证: apache-2.0 评估指标:
• 准确率 标签:
• 多模态 管道标签: 视频文本到文本 基础模型: Qwen/Qwen2-VL-7B-Instruct

💡 VideoChat-R1_7B_caption

[📂 GitHub]
[📜 技术报告]

🚀 如何使用该模型

我们提供了一个简单的安装示例：

pip install transformers
pip install qwen_vl_utils

然后您可以使用我们的模型：

from transformers import Qwen2_VLForConditionalGeneration, AutoTokenizer, AutoProcessor
from qwen_vl_utils import process_vision_info

model_path = "OpenGVLab/VideoChat-R1_7B_caption"
# 默认：在可用设备上加载模型
model = Qwen2_VLForConditionalGeneration.from_pretrained(
    model_path, torch_dtype="auto", device_map="auto",
    attn_implementation="flash_attention_2"
)

# 默认处理器
processor = AutoProcessor.from_pretrained(model_path)

video_path = "your_video.mp4"
question = "详细描述视频内容。"

messages = [
    {
        "role": "user",
        "content": [
            {
                "type": "video",
                "video": video_path,
                "max_pixels": 360 * 420,
                "fps": 1.0,
            },
            {"type": "text", "text": f""""{question} 首先在<think> </think>标签中输出思考过程，然后在<answer> </answer>标签中输出最终答案"""},
        ],
    }
]