|
--- |
|
license: apache-2.0 |
|
inference: false |
|
datasets: |
|
- Lin-Chen/ShareGPT4V |
|
pipeline_tag: image-text-to-text |
|
--- |
|
<br> |
|
<br> |
|
|
|
# open-llava-next-vicuna-7b Model Card |
|
|
|
## Model details |
|
|
|
**Model type:** |
|
open-llava-next-vicuna-7b is an open-source implementation chatbot trained by fine-tuning the entire model on open-source [Open-LLaVA-Next-mix1M](https://huggingface.co/datasets/Lin-Chen/Open-LLaVA-NeXT-mix1M) data. |
|
|
|
**Model date:** |
|
open-llava-next-vicuna-7b was trained in May 2024. |
|
|
|
**Paper or resources for more information:** |
|
[[Code](https://github.com/xiaoachen98/Open-LLaVA-NeXT)] |
|
|
|
| Name | LLM | Checkpoint | MME | SEED-image | SQA-image | MMBench | MMBench-CN | TextVQA | GQA | |
|
|---|---|---|---|---|---|---|---|---|---| |
|
| llava-next-vicuna-7b | Vicuna-7B | [HF](https://huggingface.co/liuhaotian/llava-v1.6-vicuna-7b) | 1519 | 70.2 | 70.1 | 67.4 | 60.6 | 64.9 | 64.2 | |
|
| open-llava-next-vicuna-7b | Vicuna-7B | [HF](https://huggingface.co/Lin-Chen/open-llava-next-vicuna-7b) | **1540** | **71.1** | **70.7** | **68.5** | **60.7** | **67.2** | **64.3** | |
|
|
|
## Usage |
|
|
|
You can utilize this model as we provide in our [[repository](https://github.com/xiaoachen98/Open-LLaVA-NeXT)]. Moreover, you can direct load this model and use it in the [[LLaVA repository](https://github.com/haotian-liu/LLaVA)]. |
|
|
|
## Training dataset |
|
|
|
All training data are open-sourced in our [repository](https://github.com/xiaoachen98/Open-LLaVA-NeXT). |
|
|
|
- [ShareGPT4V Mix665K](https://huggingface.co/datasets/Lin-Chen/ShareGPT4V) |
|
- 200K GPT4V-generated instruction data (ALLaVA) |
|
- 200K various VQA data |
|
|
|
## Intended use |
|
|
|
**Primary intended uses:** |
|
The primary use of open-llava-next-vicuna-7b is research on large multimodal models and chatbots. |
|
|
|
**Primary intended users:** |
|
The primary intended users of the model are researchers and hobbyists in computer vision, natural language processing, machine learning, and artificial intelligence. |
|
|