39 lines
1.4 KiB
Markdown
39 lines
1.4 KiB
Markdown
---
|
|
license: apache-2.0
|
|
base_model:
|
|
- microsoft/llava-med-v1.5-mistral-7b
|
|
- Eren-Senoglu/llava-med-v1.5-mistral-7b-hf
|
|
tags:
|
|
- medical
|
|
- vision
|
|
- llava-mistral
|
|
- text-generation
|
|
pipeline_tag: image-text-to-text
|
|
---
|
|
# 🧠 llava-med-v1.5-mistral-7b-hf
|
|
|
|
**HF-compatible conversion of [`microsoft/llava-med-v1.5-mistral-7b`](https://huggingface.co/microsoft/llava-med-v1.5-mistral-7b)**
|
|
✅ Now directly usable with [vLLM](https://github.com/vllm-project/vllm)
|
|
|
|
---
|
|
|
|
## 🔄 About This Model
|
|
|
|
This repository hosts a **Hugging Face Transformers-compatible** version of the [`microsoft/llava-med-v1.5-mistral-7b`](https://huggingface.co/microsoft/llava-med-v1.5-mistral-7b) model.
|
|
|
|
- **Original model**: Not directly usable with `vLLM` for faster inference.
|
|
- **Eren-Senoglu/llava-med-v1.5-mistral-7b-hf model**: Good, but has bug currently
|
|
- **This version**: Fully converted to the Hugging Face format and allows direct use of `vLLM`.
|
|
|
|
Thanks to this [PR](https://huggingface.co/Eren-Senoglu/llava-med-v1.5-mistral-7b-hf/discussions/1) under [Eren-Senoglu/llava-med-v1.5-mistral-7b-hf](https://huggingface.co/Eren-Senoglu/llava-med-v1.5-mistral-7b-hf) repo, I create this repo to have the fixed version.
|
|
|
|
All the contribution should go to [Eren-Senoglu](https://huggingface.co/Eren-Senoglu) and [xk-huang](https://huggingface.co/xk-huang)
|
|
|
|
## Usage
|
|
```
|
|
vllm serve wnkh/llava-med-v1.5-mistral-7b-hf
|
|
```
|
|
or
|
|
```
|
|
vllm serve wnkh/llava-med-v1.5-mistral-7b-hf --revision aeedceb
|
|
``` |