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

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

28 Dec 2023arXiv:2312.16886archive 2025-07-28

Xiangxiang Chu, Limeng Qiao, Xinyang Lin, Shuang Xu, Yang Yang, Yiming Hu, Fei Wei, Xinyu Zhang, Bo Zhang, Xiaolin Wei, Chunhua Shen

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.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

meituan-automl/mobilevlm officialmentioned in papermentioned on GitHubpytorchApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

AutoMLImage ClassificationLanguage ModelingLanguage ModellingReferring Expression ComprehensionReferring expression generation

2 archive task tags without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification ColonINST-v1 (Seen) MobileVLM-1.7B (w/ LoRA, w/ extra data) Accuray 93.64 #3 of 17 Archive leaderboard report
Image Classification ColonINST-v1 (Seen) MobileVLM-1.7B (w/o LoRA, w/ extra data) Accuray 93.02 #8 of 17 Archive leaderboard report
Image Classification ColonINST-v1 (Unseen) MobileVLM-1.7B (w/ LoRA, w/ extra data) Accuray 80.44 #3 of 17 Archive leaderboard report
Image Classification ColonINST-v1 (Unseen) MobileVLM-1.7B (w/o LoRA, w/ extra data) Accuray 78.75 #8 of 17 Archive leaderboard report
Referring expression generation ColonINST-v1 (Seen) MobileVLM-1.7B (w/ LoRA, w/ extra data) Accuray 97.87 #7 of 17 Archive leaderboard report
Referring expression generation ColonINST-v1 (Seen) MobileVLM-1.7B (w/o LoRA, w/ extra data) Accuray 97.78 #8 of 17 Archive leaderboard report
Referring expression generation ColonINST-v1 (Unseen) MobileVLM-1.7B (w/ LoRA, w/ extra data) Accuray 78.03 #2 of 17 Archive leaderboard report
Referring expression generation ColonINST-v1 (Unseen) MobileVLM-1.7B (w/o LoRA, w/ extra data) Accuray 73.14 #7 of 17 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

CLIPSETSPEED

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections