FreedomIntelligence / ALLaVA-3B

huggingface.co
Total runs: 57
24-hour runs: 0
7-day runs: 23
30-day runs: 39
Model's Last Updated: March 05 2024
text-generation

Introduction of ALLaVA-3B

Model Details of ALLaVA-3B

ALLaVA: Harnessing GPT4V-synthesized Data for A Lite Vision-Language Model

⚡ALLaVA is a project that provides a large-scale GPT4V-synthesized dataset for training LVLMs.⚡

📃 Paper • 🌐 Demo • 👨🏻‍💻 Github

🤗 ALLaVA-4V Dataset

🤗 ALLaVA-3B-Longer • 🤗 ALLaVA-3B

Benchmark Result

Our model ALLaVA-3B-Longer and ALLaVA-3B achieve competitive results on 12 benchmarks. Bold numbers denote the SOTA performance among 3B-scale models.

Model Backbone Vicuna-80 MMB SEEDBench-v1 (img) MM-Vet MMMU (val) MME TextVQA GQA EMT (CIFAR10) MLLM-Bench TouchStone LLaVA (In-the-Wild)
Qwen-VL-Chat Qwen-7B - 60.6 65.4 - 35.9 1487.5 61.5 57.5 - 6.2 711.6 -
LLaVA-v1.5-7B Vicuna-7B - 64.3 - 31.1 - 1510.7 58.2 62.0 - - 65.4
LLaVA-v1.5-13B Vicuna-13B 22.50 67.7 68.2 35.4 36.4 1531.3 61.3 63.3 85.0 7.4 637.7 70.7
ShareGPT4V-7B Vicuna-7B - 68.8 69.7 37.6 - 1943.8 60.4 63.3 - - - 72.6
TinyGPT-V Phi2-2.7B - - - - - - - 33.6 - - - -
MobileVLM MobileLLaMA-2.7B - 59.6 - - - 1288.9 47.5 - - - - -
LLaVA-Phi Phi2-2.7B - 59.8 - 28.9 - 1335.1 48.6 - - - - -
ALLaVA-3B Phi2-2.7B 48.8 64.0 65.2 32.2 35.3 1623.2 49.5 48.8 90.2 6.7 632.0 69.4
ALLaVA-3B-Longer Phi2-2.7B 52.5 64.6 65.6 35.5 33.2 1564.6 50.3 50.0 85.9 8.8 636.5 71.7

The detailed information of each benchmark is shown in Table 4 of our technical report .

🏭 Inference
Load from 🤗 (Recommended)

See the example script .

CLI

See here for CLI code snippet.

🏋️‍♂️ Training
Data
training_datasets

As shown in the table, ALLaVA-3B uses 1M and 1.5M data for PT. and FT., respectively. ALLaVA-3B-Longer trains one more epoch (i.e. 3M in total) for the FT. stage.

Code

The training code is largely based on LLaVA-v1.5 . We wholeheartedly express our gratitude for their invaluable contributions to open-sourcing LVLMs.

Cost

We train our models on 8*A800 GPUs. ALLaVA-3B-Longer takes 8.3h for PT and 21.3h for FT. ALLaVA-3B takes 8.3h for PT and 10.6h for FT. These two models share the same PT procedure.

Hyperparameters
Global Batch Size ZeRO Stage Optimizer Max LR Min LR Scheduler Max length Weight decay
256 (PT) / 128 (FT) 1 AdamW 2e-5 2e-6 CosineAnnealingWarmRestarts 2048 0

The LM backbone, projector are trainable, while the vision encoder is kept frozen. The trainabilities of each module are the same for both stages.

📚 ALLaVA-4V Data

The majority part of training data is ALLaVA-4V . See here to prepare it for training.

🙌 Contributors
📝 Citation

If you find our data useful, please consider citing our work! We are FreedomIntelligence from Shenzhen Research Institute of Big Data and The Chinese University of Hong Kong, Shenzhen

@article{chen2024allava,
  title={ALLaVA: Harnessing GPT4V-synthesized Data for A Lite Vision-Language Model},
  author={Chen, Guiming Hardy and Chen, Shunian and Zhang, Ruifei and Chen, Junying and Wu, Xiangbo and Zhang, Zhiyi and Chen, Zhihong and Li, Jianquan and Wan, Xiang and Wang, Benyou},
  journal={arXiv preprint arXiv:2402.11684},
  year={2024}
}

Runs of FreedomIntelligence ALLaVA-3B on huggingface.co

57
Total runs
0
24-hour runs
-1
3-day runs
23
7-day runs
39
30-day runs

More Information About ALLaVA-3B huggingface.co Model

More ALLaVA-3B license Visit here:

https://choosealicense.com/licenses/apache-2.0

ALLaVA-3B huggingface.co

ALLaVA-3B huggingface.co is an AI model on huggingface.co that provides ALLaVA-3B's model effect (), which can be used instantly with this FreedomIntelligence ALLaVA-3B model. huggingface.co supports a free trial of the ALLaVA-3B model, and also provides paid use of the ALLaVA-3B. Support call ALLaVA-3B model through api, including Node.js, Python, http.

FreedomIntelligence ALLaVA-3B online free

ALLaVA-3B huggingface.co is an online trial and call api platform, which integrates ALLaVA-3B's modeling effects, including api services, and provides a free online trial of ALLaVA-3B, you can try ALLaVA-3B online for free by clicking the link below.

FreedomIntelligence ALLaVA-3B online free url in huggingface.co:

https://huggingface.co/FreedomIntelligence/ALLaVA-3B

ALLaVA-3B install

ALLaVA-3B is an open source model from GitHub that offers a free installation service, and any user can find ALLaVA-3B on GitHub to install. At the same time, huggingface.co provides the effect of ALLaVA-3B install, users can directly use ALLaVA-3B installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

ALLaVA-3B install url in huggingface.co:

https://huggingface.co/FreedomIntelligence/ALLaVA-3B

Url of ALLaVA-3B

Provider of ALLaVA-3B huggingface.co

FreedomIntelligence
ORGANIZATIONS

Other API from FreedomIntelligence