--- license: mit pipeline_tag: image-text-to-text --- # Mini-InternVL-Chat-4B-V1-5 [\[📂 GitHub\]](https://github.com/OpenGVLab/InternVL) [\[🆕 Blog\]](https://internvl.github.io/blog/) [\[📜 InternVL 1.0 Paper\]](https://arxiv.org/abs/2312.14238) [\[📜 InternVL 1.5 Report\]](https://arxiv.org/abs/2404.16821) [\[🗨️ Chat Demo\]](https://internvl.opengvlab.com/) [\[🤗 HF Demo\]](https://huggingface.co/spaces/OpenGVLab/InternVL) [\[🚀 Quick Start\]](#quick-start) [\[📖 中文解读\]](https://zhuanlan.zhihu.com/p/706547971) \[🌟 [魔搭社区](https://modelscope.cn/organization/OpenGVLab) | [教程](https://mp.weixin.qq.com/s/OUaVLkxlk1zhFb1cvMCFjg) \] ## Introduction

image/png

You can run multimodal large models using a 1080Ti now. We are delighted to introduce the Mini-InternVL-Chat series. In the era of large language models, many researchers have started to focus on smaller language models, such as Gemma-2B, Qwen-1.8B, and InternLM2-1.8B. Inspired by their efforts, we have distilled our vision foundation model [InternViT-6B-448px-V1-5](https://huggingface.co/OpenGVLab/InternViT-6B-448px-V1-5) down to 300M and used [InternLM2-Chat-1.8B](https://huggingface.co/internlm/internlm2-chat-1_8b) or [Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) as our language model. This resulted in a small multimodal model with excellent performance. As shown in the figure below, we adopted the same model architecture as InternVL 1.5. We simply replaced the original InternViT-6B with InternViT-300M and InternLM2-Chat-20B with InternLM2-Chat-1.8B / Phi-3-mini-128k-instruct. For training, we used the same data as InternVL 1.5 to train this smaller model. Additionally, due to the lower training costs of smaller models, we used a context length of 8K during training. ![image/png](https://cdn-uploads.huggingface.co/production/uploads/64006c09330a45b03605bba3/rDyoe66Sqev44T0wsP5Z7.png) ## Model Details - **Model Type:** multimodal large language model (MLLM) - **Model Stats:** - Architecture: [InternViT-300M-448px](https://huggingface.co/OpenGVLab/InternViT-300M-448px) + MLP + [Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) - Image size: dynamic resolution, max to 40 tiles of 448 x 448 (4K resolution). - Params: 4.2B - **Training Strategy:** - Learnable component in the pretraining stage: MLP - Learnable component in the finetuning stage: ViT + MLP + LLM - For more details on training hyperparameters, take a look at our code: [pretrain](<>) | [finetune](<>) ## Performance ### Image Benchmarks ![image/png](https://cdn-uploads.huggingface.co/production/uploads/64119264f0f81eb569e0d569/ngl8oZvNrjItWtLUQqB2V.png) - We simultaneously use InternVL and VLMEvalKit repositories for model evaluation. Specifically, the results reported for DocVQA, ChartQA, InfoVQA, TextVQA, MME, AI2D, MMBench, CCBench, MMVet, and SEED-Image were tested using the InternVL repository. MMMU, OCRBench, RealWorldQA, HallBench, and MathVista were evaluated using the VLMEvalKit. - Please note that evaluating the same model using different testing toolkits like InternVL and VLMEvalKit can result in slight differences, which is normal. Updates to code versions and variations in environment and hardware can also cause minor discrepancies in results. - It is important to mention that the MMVet scores we report are evaluated using GPT-4-0613 as the judge model. Different versions of GPT-4 can lead to significant variations in the scores for this dataset. For instance, using GPT-4-Turbo would result in significantly lower scores. Limitations: Although we have made efforts to ensure the safety of the model during the training process and to encourage the model to generate text that complies with ethical and legal requirements, the model may still produce unexpected outputs due to its size and probabilistic generation paradigm. For example, the generated responses may contain biases, discrimination, or other harmful content. Please do not propagate such content. We are not responsible for any consequences resulting from the dissemination of harmful information. ## Quick Start We provide an example code to run Mini-InternVL-Chat-4B-V1-5 using `transformers`. > Please use transformers==4.37.2 to ensure the model works normally. ```python 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 transformers 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=6, 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=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 path = 'OpenGVLab/Mini-InternVL-Chat-4B-V1-5' model = AutoModel.from_pretrained( path, torch_dtype=torch.bfloat16, low_cpu_mem_usage=True, trust_remote_code=True).eval().cuda() tokenizer = AutoTokenizer.from_pretrained(path, trust_remote_code=True) # set the max number of tiles in `max_num` pixel_values = load_image('./examples/image1.jpg', max_num=6).to(torch.bfloat16).cuda() generation_config = dict( num_beams=1, max_new_tokens=1024, do_sample=False, ) # 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}') print(f'Assistant: {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}') print(f'Assistant: {response}') # single-image single-round conversation (单图单轮对话) question = '\nPlease describe the image shortly.' response = model.chat(tokenizer, pixel_values, question, generation_config) print(f'User: {question}') print(f'Assistant: {response}') # single-image multi-round conversation (单图多轮对话) question = '\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}') print(f'Assistant: {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}') print(f'Assistant: {response}') # multi-image multi-round conversation, combined images (多图多轮对话，拼接图像) pixel_values1 = load_image('./examples/image1.jpg', max_num=6).to(torch.bfloat16).cuda() pixel_values2 = load_image('./examples/image2.jpg', max_num=6).to(torch.bfloat16).cuda() pixel_values = torch.cat((pixel_values1, pixel_values2), dim=0) question = '\nDescribe the two images in detail.' response, history = model.chat(tokenizer, pixel_values, question, generation_config, history=None, return_history=True) 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}') print(f'Assistant: {response}') # multi-image multi-round conversation, separate images (多图多轮对话，独立图像) pixel_values1 = load_image('./examples/image1.jpg', max_num=6).to(torch.bfloat16).cuda() pixel_values2 = load_image('./examples/image2.jpg', max_num=6).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: \nImage-2: \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}') print(f'Assistant: {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}') print(f'Assistant: {response}') # batch inference, single image per sample (单图批处理) pixel_values1 = load_image('./examples/image1.jpg', max_num=6).to(torch.bfloat16).cuda() pixel_values2 = load_image('./examples/image2.jpg', max_num=6).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 = ['\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}') print(f'Assistant: {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 = './examples/red-panda.mp4' # pixel_values, num_patches_list = load_video(video_path, num_segments=32, max_num=1) 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}: \n' for i in range(len(num_patches_list))]) question = video_prefix + 'What is the red panda doing?' # Frame1: \nFrame2: \n...\nFrame31: \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}') print(f'Assistant: {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}') print(f'Assistant: {response}') ``` ## Deployment ### LMDeploy > Warning: This model is not yet supported by LMDeploy. ## License This project is released under the MIT license. ## Citation If you find this project useful in your research, please consider citing: ```BibTeX @article{chen2023internvl, title={InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks}, author={Chen, Zhe and Wu, Jiannan and Wang, Wenhai and Su, Weijie and Chen, Guo and Xing, Sen and Zhong, Muyan and Zhang, Qinglong and Zhu, Xizhou and Lu, Lewei and Li, Bin and Luo, Ping and Lu, Tong and Qiao, Yu and Dai, Jifeng}, journal={arXiv preprint arXiv:2312.14238}, year={2023} } @article{chen2024far, title={How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites}, author={Chen, Zhe and Wang, Weiyun and Tian, Hao and Ye, Shenglong and Gao, Zhangwei and Cui, Erfei and Tong, Wenwen and Hu, Kongzhi and Luo, Jiapeng and Ma, Zheng and others}, journal={arXiv preprint arXiv:2404.16821}, year={2024} } ```