File size: 1,059 Bytes
581a501 ced1458 581a501 ced1458 bc4a97b ced1458 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
---
license: apache-2.0
tags:
- MobileVLM V2
---
## Model Summery
MobileVLM V2 is a family of significantly improved vision language models upon MobileVLM, which proves that a delicate orchestration of novel architectural design, an improved training scheme tailored for mobile VLMs, and rich high-quality dataset curation can substantially benefit VLMs’ performance. Specifically, MobileVLM V2 1.7B achieves better or on-par performance on standard VLM benchmarks compared with much larger VLMs at the 3B scale. Notably, MobileVLM_V2-3B model outperforms a large variety of VLMs at the 7B+ scale.
The MobileVLM_V2-7B was built on [Vicuna-7B-v1.5](https://huggingface.co/lmsys/vicuna-7b-v1.5) to facilitate the off-the-shelf deployment.
## Model Sources
- Repository: https://github.com/Meituan-AutoML/MobileVLM
- Paper: [MobileVLM V2: Faster and Stronger Baseline for Vision Language Model](https://arxiv.org/abs/2402.03766)
## How to Get Started with the Model
Inference examples can be found at [Github](https://github.com/Meituan-AutoML/MobileVLM).
|