metadata

license: mit
datasets:
  - laion/laion2B-en
  - laion/laion-coco
  - laion/laion2B-multi
  - kakaobrain/coyo-700m
  - conceptual_captions
  - wanng/wukong100m

Model Card for InternVL-Chat-Chinese-V1.2

What is InternVL?

[Paper] [GitHub] [Chat Demo]

InternVL scales up the ViT to 6B parameters and aligns it with LLM.

InternVL-Chat-V1.2 Blog

Date: 2024/02/12
Developed by: Zhe Chen, Weiyun Wang, Wenhai Wang, Erfei Cui, Zhangwei Gao, Xizhou Zhu, Lewei Lu, Tong Lu, Yu Qiao, Jifeng Dai

We are excited to introduce InternVL-Chat-V1.2. Inspired by LLaVA-NeXT-34B, we have also adopted Nous-Hermes-2-Yi-34B as the language model. Below is the pipeline.

From the experimental results, we've observed that a stronger language model (34B) can better leverage the powerful capabilities of our vision foundation model (InternViT-6B).

For better training reproducibility, we follow the minimalist design and data efficiency similar to LLaVA-NeXT. To reduce training costs, we provide a pre-trained MLP projector and only employ around 1 million visual instruction tuning samples for SFT. Our model has a total of 40 billion parameters and can be trained within 1.5 days using 32 A100 GPUs. The code, data, and model will be made publicly available.

Data Preparation

Inspired by LLaVA-NeXT, we adopted a data-efficient SFT strategy to train InternVL-Chat-V1.2, utilizing approximately 1.2M of visual instruction tuning samples in total, all of which are fully open-source. In a macro sense, we build upon ShareGPT-4V and additionally integrate LLaVA-ZH, DVQA, ChartQA, AI2D, DocVQA, GeoQA+, and SynthDoG-EN. Most of the data remains consistent with LLaVA-NeXT.

For more details about data preparation, please see here.

Performance

* Proprietary Model

name	image size	MMMU (val)	MMMU (test)	MathVista (testmini)	MMB (test)	MMB−CN (test)	MMVP	MME	ScienceQA (image)	POPE	TextVQA	SEEDv1 (image)	VizWiz (test)	GQA (test)
GPT-4V*	unknown	56.8	55.7	49.9	77.0	74.4	38.7	1409/517	-	-	78.0	71.6	-	-
Gemini Ultra*	unknown	59.4	-	53.0	-	-	-	-	-	-	82.3	-	-	-
Gemini Pro*	unknown	47.9	-	45.2	73.6	74.3	40.7	1497/437	-	-	74.6	70.7	-	-
Qwen-VL-Plus*	unknown	45.2	40.8	43.3	67.0	70.7	-	1681/502	-	-	78.9	65.7	-	-
Qwen-VL-Max*	unknown	51.4	46.8	51.0	77.6	75.7	-	-	-	-	79.5	-	-	-

LLaVA-NEXT-34B	672x672	51.1	44.7	46.5	79.3	79.0	-	1631/397	81.8	87.7	69.5	75.9	63.8	67.1
InternVL-Chat-V1.2	448x448	51.6	46.2	47.7	82.2	81.2	56.7	1672/509	83.3	88.0	69.7	75.6	60.0	64.0

MMBench results are collected from the leaderboard.
In most benchmarks, InternVL-Chat-V1.2 achieves better performance than LLaVA-NeXT-34B.

Training (SFT)

We provide slurm scripts for multi-node multi-GPU training. You can use either 32 or 64 GPUs to train this model. If you use 64 GPUs, training will take approximately 18 hours.

For more details about training, please see here.

The hyperparameters used for finetuning are listed in the following table.

Hyperparameter	Trainable Param	Global Batch Size	Learning rate	Epochs	Max length	Weight decay
InternVL-Chat-V1.2	40B (full model)	512	1e-5	1	2048	0.05

Model Details

Model Type: vision large language model, multimodal chatbot
Model Stats:
- Architecture: InternViT-6B-448px-V1-2 + MLP + Nous-Hermes-2-Yi-34B
- Params: 40B
- Image size: 448 x 448
- Number of visual tokens: 256
Training Strategy:
- Pretraining Stage
  - Learnable Component: MLP
  - Data: Trained on 8192x4800=39.3M samples, including COYO, LAION, CC12M, CC3M, SBU, Wukong, GRIT, Objects365, OpenImages, and OCR data.
  - Note: In this stage, we load the pretrained weights of InternViT-6B-448px-V1-2. Moreover, in order to reduce the number of visual tokens, we use a pixel shuffle to reduce 1024 tokens to 256 tokens.
- SFT Stage
  - Learnable Component: ViT + MLP + LLM
  - Data: A simplified, fully open-source dataset, containing approximately 1 million samples.

Model Usage

We provide a minimum code example to run InternVL-Chat using only the transformers library.

You also can use our online demo for a quick experience of this model.

Note: If you meet this error ImportError: This modeling file requires the following packages that were not found in your environment: fastchat, please run pip install fschat.

import torch
from PIL import Image
from transformers import AutoModel, CLIPImageProcessor
from transformers import AutoTokenizer

path = "OpenGVLab/InternVL-Chat-Chinese-V1-2"
model = AutoModel.from_pretrained(
    path,
    torch_dtype=torch.bfloat16,
    low_cpu_mem_usage=True,
    trust_remote_code=True,
    device_map='auto').eval()

tokenizer = AutoTokenizer.from_pretrained(path)
image = Image.open('./examples/image2.jpg').convert('RGB')
image = image.resize((448, 448))
image_processor = CLIPImageProcessor.from_pretrained(path)

pixel_values = image_processor(images=image, return_tensors='pt').pixel_values
pixel_values = pixel_values.to(torch.bfloat16).cuda()

generation_config = dict(
    num_beams=1,
    max_new_tokens=512,
    do_sample=False,
)

question = "请详细描述图片"
response = model.chat(tokenizer, pixel_values, question, generation_config)

Citation

If you find this project useful in your research, please consider citing:

@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}
}

License

This project is released under the MIT license. Parts of this project contain code and models (e.g., LLaMA2) from other sources, which are subject to their respective licenses.

Acknowledgement

InternVL is built with reference to the code of the following projects: OpenAI CLIP, Open CLIP, CLIP Benchmark, EVA, InternImage, ViT-Adapter, MMSegmentation, Transformers, DINOv2, BLIP-2, Qwen-VL, and LLaVA-1.5. Thanks for their awesome work!