MobileVLM : A Fast, Strong and Open Vision Language Assistant for Mobile Devices

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chu, Xiangxiang, Qiao, Limeng, Lin, Xinyang, Xu, Shuang, Yang, Yang, Hu, Yiming, Wei, Fei, Zhang, Xinyu, Zhang, Bo, Wei, Xiaolin, Shen, Chunhua
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914625738506240
author Chu, Xiangxiang
Qiao, Limeng
Lin, Xinyang
Xu, Shuang
Yang, Yang
Hu, Yiming
Wei, Fei
Zhang, Xinyu
Zhang, Bo
Wei, Xiaolin
Shen, Chunhua
author_facet Chu, Xiangxiang
Qiao, Limeng
Lin, Xinyang
Xu, Shuang
Yang, Yang
Hu, Yiming
Wei, Fei
Zhang, Xinyu
Zhang, Bo
Wei, Xiaolin
Shen, Chunhua
contents We present MobileVLM, a competent multimodal vision language model (MMVLM) targeted to run on mobile devices. It is an amalgamation of a myriad of architectural designs and techniques that are mobile-oriented, which comprises a set of language models at the scale of 1.4B and 2.7B parameters, trained from scratch, a multimodal vision model that is pre-trained in the CLIP fashion, cross-modality interaction via an efficient projector. We evaluate MobileVLM on several typical VLM benchmarks. Our models demonstrate on par performance compared with a few much larger models. More importantly, we measure the inference speed on both a Qualcomm Snapdragon 888 CPU and an NVIDIA Jeston Orin GPU, and we obtain state-of-the-art performance of 21.5 tokens and 65.3 tokens per second, respectively. Our code will be made available at: https://github.com/Meituan-AutoML/MobileVLM.
format Preprint
id arxiv_https___arxiv_org_abs_2312_16886
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle MobileVLM : A Fast, Strong and Open Vision Language Assistant for Mobile Devices
Chu, Xiangxiang
Qiao, Limeng
Lin, Xinyang
Xu, Shuang
Yang, Yang
Hu, Yiming
Wei, Fei
Zhang, Xinyu
Zhang, Bo
Wei, Xiaolin
Shen, Chunhua
Computer Vision and Pattern Recognition
We present MobileVLM, a competent multimodal vision language model (MMVLM) targeted to run on mobile devices. It is an amalgamation of a myriad of architectural designs and techniques that are mobile-oriented, which comprises a set of language models at the scale of 1.4B and 2.7B parameters, trained from scratch, a multimodal vision model that is pre-trained in the CLIP fashion, cross-modality interaction via an efficient projector. We evaluate MobileVLM on several typical VLM benchmarks. Our models demonstrate on par performance compared with a few much larger models. More importantly, we measure the inference speed on both a Qualcomm Snapdragon 888 CPU and an NVIDIA Jeston Orin GPU, and we obtain state-of-the-art performance of 21.5 tokens and 65.3 tokens per second, respectively. Our code will be made available at: https://github.com/Meituan-AutoML/MobileVLM.
title MobileVLM : A Fast, Strong and Open Vision Language Assistant for Mobile Devices
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.16886