pipeline_tag: 图像文本到文本 library_name: transformers license: apache-2.0

ChemVLM-8B：面向化学领域的多模态大语言模型

这是ChemVLM的80亿参数版本，专为化学应用设计的跨模态大语言模型。

论文

ChemVLM：探索多模态大语言模型在化学领域的能力

摘要

大语言模型（LLMs）已取得显著成功，并应用于包括化学在内的多个科学领域。然而，许多化学任务需要处理视觉信息，现有化学大语言模型无法有效应对。这促使化学领域对能整合多模态信息模型的需求日益增长。本文介绍ChemVLM，一个专为化学应用设计的开源多模态大语言模型。ChemVLM基于精心构建的双语多模态数据集训练，增强了其理解文本和视觉化学信息的能力，包括分子结构、化学反应和化学试题。我们开发了三个评估数据集，分别针对化学光学字符识别（OCR）、多模态化学推理（MMCR）和多模态分子理解任务。我们在多项任务中将ChemVLM与开源和商业多模态大语言模型进行对比测试。实验结果表明，ChemVLM在所有评估任务中均展现出竞争力。模型地址：https://huggingface.co/AI4Chem/ChemVLM-26B。

模型描述

ChemVLM架构基于InternVLM，整合了视觉与语言处理组件。该模型在包含分子结构、化学反应和化学试题的双语多模态化学数据集上训练。更多架构细节详见Github说明文档。

引用

@inproceedings{li2025chemvlm,
  title={Chemvlm: 探索多模态大语言模型在化学领域的能力},
  author={李俊贤 and 张迪 and 王勋智 and 郝泽颖 and 雷静迪 and 谭茜 and 周才 and 刘伟 and 杨耀天 and 熊昕睿等},
  booktitle={AAAI人工智能会议论文集},
  volume={39},
  number={1},
  pages={415--423},
  year={2025}
}

代码库与数据集

代码库及数据集详见：https://github.com/AI4Chem/ChemVlm。

80亿参数模型在多项任务中的表现

快速开始

from transformers import AutoTokenizer, AutoModelforCasualLM
import torch
import torchvision.transforms as T
import transformers
from torchvision.transforms.functional import InterpolationMode


IMAGENET_MEAN = (0.485, 0.456, 0.406)
IMAGENET_STD = (0.229, 0.224, 0.225)

def build_transform(input_size):
    MEAN, STD = IMAGENET_MEAN, IMAGENET_STD
    transform = T.Compose([
        T.Lambda(lambda img: img.convert('RGB') if img.mode != 'RGB' else img),
        T.Resize((input_size, input_size), interpolation=InterpolationMode.BICUBIC),
        T.ToTensor(),
        T.Normalize(mean=MEAN, std=STD)
    ])
    return transform


def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):
    best_ratio_diff = float('inf')
    best_ratio = (1, 1)
    area = width * height
    for ratio in target_ratios:
        target_aspect_ratio = ratio[0] / ratio[1]
        ratio_diff = abs(aspect_ratio - target_aspect_ratio)
        if ratio_diff < best_ratio_diff:
            best_ratio_diff = ratio_diff
            best_ratio = ratio
        elif ratio_diff == best_ratio_diff:
            if area > 0.5 * image_size * image_size * ratio[0] * ratio[1]:
                best_ratio = ratio
    return best_ratio


def dynamic_preprocess(image, min_num=1, max_num=6, image_size=448, use_thumbnail=False):
    orig_width, orig_height = image.size
    aspect_ratio = orig_width / orig_height

    target_ratios = set(
        (i, j) for n in range(min_num, max_num + 1) for i in range(1, n + 1) for j in range(1, n + 1) if
        i * j <= max_num and i * j >= min_num)
    target_ratios = sorted(target_ratios, key=lambda x: x[0] * x[1])

    target_aspect_ratio = find_closest_aspect_ratio(
        aspect_ratio, target_ratios, orig_width, orig_height, image_size)

    target_width = image_size * target_aspect_ratio[0]
    target_height = image_size * target_aspect_ratio[1]
    blocks = target_aspect_ratio[0] * target_aspect_ratio[1]

    resized_img = image.resize((target_width, target_height))
    processed_images = []
    for i in range(blocks):
        box = (
            (i % (target_width // image_size)) * image_size,
            (i // (target_width // image_size)) * image_size,
            ((i % (target_width // image_size)) + 1) * image_size,
            ((i // (target_width // image_size)) + 1) * image_size
        )
        split_img = resized_img.crop(box)
        processed_images.append(split_img)
    assert len(processed_images) == blocks
    if use_thumbnail and len(processed_images) != 1:
        thumbnail_img = image.resize((image_size, image_size))
        processed_images.append(thumbnail_img)
    return processed_images


def load_image(image_file, input_size=448, max_num=6):
    image = Image.open(image_file).convert('RGB')
    transform = build_transform(input_size=input_size)
    images = dynamic_preprocess(image, image_size=input_size, use_thumbnail=True, max_num=max_num)
    pixel_values = [transform(image) for image in images]
    pixel_values = torch.stack(pixel_values)
    return pixel_values


tokenizer = AutoTokenizer.from_pretrained('AI4Chem/ChemVLM-8B', trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    "AI4Chem/ChemVLM-8B",
    torch_dtype=torch.bfloat16,
    low_cpu_mem_usage=True,
    trust_remote_code=True
).cuda().eval()

query = "请描述图片中的分子结构"
image_path = "图片路径"
pixel_values = load_image(image_path, max_num=6).to(torch.bfloat16).cuda()

gen_kwargs = {"max_length": 1000, "do_sample": True, "temperature": 0.7, "top_p": 0.9}

response = model.chat(tokenizer, pixel_values, query, gen_kwargs)
print(response)

安装依赖库：pip install transformers>=4.37.0 sentencepiece einops timm accelerate>=0.26.0。请确保已安装torch和torchvision。