🙋魔搭ModelScope本期社区进展：

📟1338个模型：HunyuanVideo、InternVL2.5系列、TeleChat2系列、AWPortraitCN、fish-speech-1.5、Mistral-Nemo-Instruct-2407-FP等；

📁82个数据集：OpenO1-SFT、LLaVA-CoT-o1-Instruct、LLaVA-CoT-100k等；

🎨26个创新应用：OminiControl、Hallo、HunyuanVideo等；

📄 8篇内容：

  • NexaAI, 一行命令运行魔搭社区模型，首次在设备上运行 Qwen2-Audio

  • 魔搭AIGC12月赛题公布&11月获奖作品出炉

  • 开发者福利，魔搭推出免费模型推理API，注册就送每日2000次调用！

  • 2024 “AI+硬件创新大赛”获奖名单出炉，浙大、上交与复旦联队等夺冠

  • 腾讯开源混元视频生成模型，这效果！太稳了吧！

  • 金融行业 · 大模型挑战赛 ｜用大模型理解金融市场

  • 智源研究院发布中文高质量数据集CCI3.0-HQ技术报告

  • 用 OpenVINO™ 部署 GLM-Edge 全家桶

01.精选模型

HunyuanVideo

HunyuanVideo是腾讯开源的一个具有超过130亿参数的先进开源视频生成模型，它通过集成数据管理、图像-视频联合训练和高效基础设施的综合框架，以及有效的模型架构和数据集扩展策略，实现了高视觉质量、运动多样性、文本-视频对齐和生成稳定性。

模型链接：

https://modelscope.cn/models/AI-ModelScope/HunyuanVideo

代码示例：

详见 腾讯开源混元视频生成模型，这效果！太稳了吧！

 

InternVL2.5系列

InternVL 2.5是上海人工智能实验室开源的一个先进的多模态大型语言模型（MLLM）系列，基于InternVL 2.0的核心模型架构，在训练、测试策略以及数据质量方面作显著增强，通过对包括多学科推理、文档理解、多图像/视频理解、现实世界理解、多模态幻觉检测、视觉锚定、多语言能力和纯语言处理等广泛基准的深入评估，是第一个在MMMU基准上超过70%的开源MLLM。

模型链接：

  • InternVL2_5-1B/summary：

https://www.modelscope.cn/models/OpenGVLab/InternVL2_5-1B/summary

 

  • InternVL2_5-2B：

https://www.modelscope.cn/models/OpenGVLab/InternVL2_5-2B

  • InternVL2_5-4B：

https://www.modelscope.cn/models/OpenGVLab/InternVL2_5-4B

 

  • InternVL2_5-8B：

https://www.modelscope.cn/models/OpenGVLab/InternVL2_5-8B

 

  • InternVL2_5-26B：

https://www.modelscope.cn/models/OpenGVLab/InternVL2_5-26B

  • InternVL2_5-38B：

https://www.modelscope.cn/models/OpenGVLab/InternVL2_5-38B

 

  • InternVL2_5-78B：

https://www.modelscope.cn/models/OpenGVLab/InternVL2_5-78B

代码示例：

transformers推理代码（以InternVL2_5-4B推理为例）：


import numpy as np
import torch
import torchvision.transforms as T
from decord import VideoReader, cpu
from PIL import Image
from torchvision.transforms.functional import InterpolationMode
from modelscope import AutoModel, AutoTokenizer

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=12, image_size=448, use_thumbnail=False):
    orig_width, orig_height = image.size
    aspect_ratio = orig_width / orig_height

    # calculate the existing image aspect ratio
    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])

    # find the closest aspect ratio to the target
    target_aspect_ratio = find_closest_aspect_ratio(
        aspect_ratio, target_ratios, orig_width, orig_height, image_size)

    # calculate the target width and height
    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]

    # resize the image
    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 the image
        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=12):
    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

# If you want to load a model using multiple GPUs, please refer to the `Multiple GPUs` section.
path = 'OpenGVLab/InternVL2_5-4B'
model = AutoModel.from_pretrained(
    path,
    torch_dtype=torch.bfloat16,
    low_cpu_mem_usage=True,
    use_flash_attn=True,
    trust_remote_code=True).eval().cuda()
tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code=True, use_fast=False)

# set the max number of tiles in `max_num`
pixel_values = load_image('./awesome.png', max_num=12).to(torch.bfloat16).cuda()
generation_config = dict(max_new_tokens=1024, do_sample=True)

# pure-text conversation (纯文本对话)
question = 'Hello, who are you?'
response, history = model.chat(tokenizer, None, question, generation_config, history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'Can you tell me a story?'
response, history = model.chat(tokenizer, None, question, generation_config, history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

# single-image single-round conversation (单图单轮对话)
question = '<image>\nPlease describe the image shortly.'
response = model.chat(tokenizer, pixel_values, question, generation_config)
print(f'User: {question}\nAssistant: {response}')

# single-image multi-round conversation (单图多轮对话)
question = '<image>\nPlease describe the image in detail.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config, history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'Please write a poem according to the image.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config, history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

# multi-image multi-round conversation, combined images (多图多轮对话，拼接图像)
pixel_values1 = load_image('./awesome.png', max_num=12).to(torch.bfloat16).cuda()
pixel_values2 = load_image('./noword.jpg', max_num=12).to(torch.bfloat16).cuda()
pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)

question = '<image>\nDescribe the two images in detail.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'What are the similarities and differences between these two images.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

# multi-image multi-round conversation, separate images (多图多轮对话，独立图像)
pixel_values1 = load_image('./awesome.png', max_num=12).to(torch.bfloat16).cuda()
pixel_values2 = load_image('./noword.jpg', max_num=12).to(torch.bfloat16).cuda()
pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)
num_patches_list = [pixel_values1.size(0), pixel_values2.size(0)]

question = 'Image-1: <image>\nImage-2: <image>\nDescribe the two images in detail.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               num_patches_list=num_patches_list,
                               history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'What are the similarities and differences between these two images.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               num_patches_list=num_patches_list,
                               history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

# batch inference, single image per sample (单图批处理)
pixel_values1 = load_image('./awesome.png', max_num=12).to(torch.bfloat16).cuda()
pixel_values2 = load_image('./noword.jpg', max_num=12).to(torch.bfloat16).cuda()
num_patches_list = [pixel_values1.size(0), pixel_values2.size(0)]
pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0)

questions = ['<image>\nDescribe the image in detail.'] * len(num_patches_list)
responses = model.batch_chat(tokenizer, pixel_values,
                             num_patches_list=num_patches_list,
                             questions=questions,
                             generation_config=generation_config)
for question, response in zip(questions, responses):
    print(f'User: {question}\nAssistant: {response}')

# video multi-round conversation (视频多轮对话)
def get_index(bound, fps, max_frame, first_idx=0, num_segments=32):
    if bound:
        start, end = bound[0], bound[1]
    else:
        start, end = -100000, 100000
    start_idx = max(first_idx, round(start * fps))
    end_idx = min(round(end * fps), max_frame)
    seg_size = float(end_idx - start_idx) / num_segments
    frame_indices = np.array([
        int(start_idx + (seg_size / 2) + np.round(seg_size * idx))
        for idx in range(num_segments)
    ])
    return frame_indices

def load_video(video_path, bound=None, input_size=448, max_num=1, num_segments=32):
    vr = VideoReader(video_path, ctx=cpu(0), num_threads=1)
    max_frame = len(vr) - 1
    fps = float(vr.get_avg_fps())

    pixel_values_list, num_patches_list = [], []
    transform = build_transform(input_size=input_size)
    frame_indices = get_index(bound, fps, max_frame, first_idx=0, num_segments=num_segments)
    for frame_index in frame_indices:
        img = Image.fromarray(vr[frame_index].asnumpy()).convert('RGB')
        img = dynamic_preprocess(img, image_size=input_size, use_thumbnail=True, max_num=max_num)
        pixel_values = [transform(tile) for tile in img]
        pixel_values = torch.stack(pixel_values)
        num_patches_list.append(pixel_values.shape[0])
        pixel_values_list.append(pixel_values)
    pixel_values = torch.cat(pixel_values_list)
    return pixel_values, num_patches_list

video_path = './showcase.mp4'
pixel_values, num_patches_list = load_video(video_path, num_segments=8, max_num=1)
pixel_values = pixel_values.to(torch.bfloat16).cuda()
video_prefix = ''.join([f'Frame{i+1}: <image>\n' for i in range(len(num_patches_list))])
question = video_prefix + 'What is the red panda doing?'
# Frame1: <image>\nFrame2: <image>\n...\nFrame8: <image>\n{question}
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               num_patches_list=num_patches_list, history=None, return_history=True)
print(f'User: {question}\nAssistant: {response}')

question = 'Describe this video in detail. Don\'t repeat.'
response, history = model.chat(tokenizer, pixel_values, question, generation_config,
                               num_patches_list=num_patches_list, history=history, return_history=True)
print(f'User: {question}\nAssistant: {response}')

 

TeleChat2

星辰语义大模型TeleChat2是由中国电信人工智能研究院研发训练的大语言模型。

TeleChat2系列模型包括TeleChat2-3B、TeleChat2-7B、TeleChat2-35B和TeleChat2-115B，具备显著的工具调用功能和优化的Function Call表现，相较同尺寸模型在多个榜单中表现优异。

该系列在训练数据和方法方面进行了全面改进，通用问答及逻辑推理等能力较前代提升超过29%。所有模型均在国产算力和深度学习框架上训练，保障了自主可控性。通过优化MP、PP、SP实现和长文训练技术，提升了模型性能和训练速度。

模型链接：

  • telechat2-3B

https://modelscope.cn/models/TeleAI/TeleChat2-3B

  • telechat2-7B

https://modelscope.cn/models/TeleAI/TeleChat2-7B

 

  • telechat2-35B

https://modelscope.cn/models/TeleAI/TeleChat2-35B-Nov

  • telechat2-115B

https://modelscope.cn/models/TeleAI/TeleChat2-115B

代码示例：

当前模型推理兼容了单卡和多卡推理，以及针对长文推理做了部分优化工作，模型推理方法示范：


import os
import torch
from modelscope import AutoModelForCausalLM, AutoTokenizer, GenerationConfig
tokenizer = AutoTokenizer.from_pretrained('TeleAI/TeleChat2-3B', trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained('TeleAI/TeleChat2-3B', trust_remote_code=True, device_map="auto",
                                                  torch_dtype=torch.float16)
prompt = "生抽与老抽的区别？"
messages = [{"role": "user", "content": prompt}]
text = tokenizer.apply_chat_template(messages,
  tokenize=False,
       add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

AWPortraitCN

AWPortraitCN 是基于FLUX.1-dev开发。它在训练时使用了更符合中国人外貌和审美特征的图像。它包含了多种类型的肖像，如室内和室外肖像、时尚和影棚照片。它具有很强的泛化能力。与原始版本相比，AWPortraitCN在皮肤质感上更加细腻真实。为了追求更真实的原始图像效果，它可以与AWPortraitSR工作流程一起使用。

 

模型链接：

https://modelscope.cn/models/LiblibAI/AWPortraitCN

 

效果展示：

 

02.数据集推荐

DAVIS-Edit

数据集专为通过SFT微调语言模型以激活思维链而设计，目的是提升模型在复杂推理任务上生成连贯、逻辑性强的推理序列的能力。

 

数据集链接：

https://modelscope.cn/datasets/AI-ModelScope/OpenO1-SFT

 

LLaVA-CoT-o1-Instruct

LLaVA-CoT-o1-Instruct是用于支持和提升语言模型在复杂任务中的逻辑推理和连贯性表现。

 

数据集链接：

https://modelscope.cn/datasets/AI-ModelScope/LLaVA-CoT-o1-Instruct

 

LLaVA-CoT-100k

LLaVA-CoT-100k是一个专注于复杂推理任务的数据集，包含100,000个样本。这个数据集被设计用来训练和测试语言模型，以提高它们在逻辑推理和问题解决方面的能力。

数据集链接：

https://modelscope.cn/datasets/AI-ModelScope/LLaVA-CoT-100k

 

03.精选应用

OminiControl

是一款内容生成工具，支持上传一张有主体的图片+一段文本关键词进行合成。具有广泛的应用前景。

体验直达：

https://modelscope.cn/studios/AI-ModelScope/OminiControl


Hallo 

TTS x Hallo Talking Portrait Generator是一个集成多个开源项目的演示工具，允许用户生成会说话的肖像视频，但视频长度限制为4秒音频以内。

 

体验直达：

https://modelscope.cn/studios/AI-ModelScope/Hallo

HunyuanVideo

HunyuanVideo是一个创新的视频生成工具，它通过引入Transformer架构和Full Attention机制，实现了图像和视频的统一生成。该工具采用“Dual-stream to Single-stream”混合模型设计，首先独立处理视频和文本token，确保每个模态都能学习到适合自己的调制机制，然后在Single-stream阶段将视频和文本token融合，实现多模态信息的有效整合，捕捉视觉和语义信息之间的复杂交互，从而显著提升模型的整体性能。

体验直达：

https://modelscope.cn/studios/chuanSir/HunyuanVideo

 

04.社区精选文章

 

 

 

 

Logo
欢迎加入ModelScope魔搭中文开源社区

ModelScope旨在打造下一代开源的模型即服务共享平台，为泛AI开发者提供灵活、易用、低成本的一站式模型服务产品，让模型应用更简单！

更多推荐

AI赋能大学计划·大模型技术与应用实战学生训练营——电子科技大学站圆满结营

12月05日，由中国软件行业校园招聘与实习公共服务平台携手阿里魔搭社区共同举办的AI赋能大学计划·大模型技术与产业趋势高校行AIGC项目实战营·电子科技大学站圆满结营。

avatar ModelScope魔搭社区
cover

Llama 3.3开源！70B媲美405B性能，支持128K上下文

avatar ModelScope魔搭社区
cover

InternVL 2.5，首个MMMU超过70%的开源模型，性能媲美GPT-4o

avatar ModelScope魔搭社区