Papers › MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer

MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer

5 Oct 2021ICLR 2022 4arXiv:2110.02178archive 2025-07-28

Sachin Mehta, Mohammad Rastegari

Light-weight convolutional neural networks (CNNs) are the de-facto for mobile vision tasks. Their spatial inductive biases allow them to learn representations with fewer parameters across different vision tasks. However, these networks are spatially local. To learn global representations, self-attention-based vision trans-formers (ViTs) have been adopted. Unlike CNNs, ViTs are heavy-weight. In this paper, we ask the following question: is it possible to combine the strengths of CNNs and ViTs to build a light-weight and low latency network for mobile vision tasks? Towards this end, we introduce MobileViT, a light-weight and general-purpose vision transformer for mobile devices. MobileViT presents a different perspective for the global processing of information with transformers, i.e., transformers as convolutions. Our results show that MobileViT significantly outperforms CNN- and ViT-based networks across different tasks and datasets. On the ImageNet-1k dataset, MobileViT achieves top-1 accuracy of 78.4% with about 6 million parameters, which is 3.2% and 6.2% more accurate than MobileNetv3 (CNN-based) and DeIT (ViT-based) for a similar number of parameters. On the MS-COCO object detection task, MobileViT is 5.7% more accurate than MobileNetv3 for a similar number of parameters. Our source code is open-source and available at: https://github.com/apple/ml-cvnets

PaperPDFConference PDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2110.02178")

Code

Syntology Ran 53 of 68 code samples harvested from 9 repositories linked to this paper; 15 have no recorded run. Of those that ran: 1 ran · honoured contract; 9 ran · our draft was wrong; 43 ran with no contract checked.

By repository: community (archive-listed): 66 samples from 9 repositories, 51 ran; 2 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

31 repositories listed; official and paper-mentioned ones first.

apple/ml-cvnets officialmentioned in papermentioned on GitHubpytorchNOASSERTION report
IMvision12/keras-vision-models mentioned on GitHubpytorch report
chinhsuanwu/mobilevit-pytorch mentioned on GitHubpytorch report
godhj93/MobileViT mentioned on GitHubtf report
hkzhang91/edgeformer mentioned on GitHubpytorch report
hkzhang91/pacc-net mentioned on GitHubpytorch report
hkzhang91/parc-net mentioned on GitHubpytorch report
jaiwei98/mobile-vit-pytorch mentioned on GitHubpytorch report
james77777778/keras-image-models mentioned on GitHubpytorchApache-2.0 report
microndla/mobilevitv3 mentioned on GitHubpytorchNOASSERTION report
mmaaz60/EdgeNeXt mentioned on GitHubpytorch report
murufeng/awesome_lightweight_networks mentioned on GitHubpytorch report
roydenwa/cell-centroid-former mentioned on GitHubtf report
rwightman/pytorch-image-models mentioned on GitHubpytorch report
t0nyliang/EEGMobile mentioned on GitHubpytorch report
wilile26811249/MobileViT mentioned on GitHubpytorch report
PaddlePaddle/PaddleClas paddleApache-2.0 report
pwc-1/Paper-5 mindspore report
pwc-1/Paper-9 mindspore 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

68 samples harvested; 53 ran; 1 honoured the contract we drafted; 15 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
9ran · our draft was wrong
43ran
15unverified

Licence: 18 of the 68 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 9 repositories linked to this paper, official or community; each sample names its own and says which. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

Attention jaiwei98/mobile-vit-pytorch/mobile_vit/mobilevit.py community (archive-listed) ran · metamorphic tier: deterministic no licence file found · pointer only · 7dce48772b2e159a · report
Attention murufeng/awesome_lightweight_networks/light_cnns/Transformer/mobile_vit.py community (archive-listed) ran · metamorphic tier: invariant MIT (permissive) · 43531c7c47f675ef · report
Attention xmu-xiaoma666/External-Attention-pytorch/model/backbone/MobileViT.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · d55210e5cfb23cea · report
Attention kornia/kornia/kornia/models/vit_mobile.py community (archive-listed) ran · metamorphic tier: invariant Apache-2.0 (permissive) · 2230f658d6c25822 · report
ChannelMLP roydenwa/cell-centroid-former/src/cellcentroidformer/model_architecture.py community (archive-listed) ran no licence file found · pointer only · f1f1c5ff227fe41d · report
ConvEncoder mmaaz60/EdgeNeXt/models/edgenext.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · a75100d19286cb15 · report
ConvNormAct wilile26811249/MobileViT/models/model.py community (archive-listed) ran fingerprinted MIT (permissive) · 8ed7073f6f37a1a1 · report
Conv_BN_ReLU murufeng/awesome_lightweight_networks/light_cnns/Transformer/mobile_vit.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 38ab7af20618ac3e · report
EncoderLayer godhj93/MobileViT/utils/nets/MobileViT.py community (archive-listed) ran no licence file found · pointer only · 20ec52f2f7242b3d · report
FFN wilile26811249/MobileViT/models/model.py community (archive-listed) ran MIT (permissive) · 0700a6a34626a6bf · report
FeedForward murufeng/awesome_lightweight_networks/light_cnns/Transformer/mobile_vit.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 4d1e816536427160 · report
FeedForward xmu-xiaoma666/External-Attention-pytorch/model/backbone/MobileViT.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 878fe3f52adf2d34 · report
FeedForward chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) ran · metamorphic tier: invariant MIT (permissive) · 58080bbc3d99ba0f · report
FeedForward kornia/kornia/kornia/models/vit_mobile.py community (archive-listed) ran · metamorphic tier: invariant Apache-2.0 (permissive) · 144b7a438af242ef · report
InvertedResidual jaiwei98/mobile-vit-pytorch/mobile_vit/mobilevit.py community (archive-listed) ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · c56c074407924b8b · report
InvertedResidual wilile26811249/MobileViT/models/model.py community (archive-listed) ran fingerprinted MIT (permissive) · cc6fad7bef0b8c04 · report
LayerNorm mmaaz60/EdgeNeXt/models/edgenext.py community (archive-listed) ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · d59faa990dd0baa0 · report
MV2Block murufeng/awesome_lightweight_networks/light_cnns/Transformer/mobile_vit.py community (archive-listed) ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · e789209b847f99a6 · report
MV2Block xmu-xiaoma666/External-Attention-pytorch/model/backbone/MobileViT.py community (archive-listed) ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 7a5ed61c85d652d3 · report
MV2Block chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) ran · metamorphic tier: invariant MIT (permissive) · d8b589747023f0ed · report
MV2Block kornia/kornia/kornia/models/vit_mobile.py community (archive-listed) ran · metamorphic tier: invariant fingerprinted Apache-2.0 (permissive) · ed0efa372d4badcc · report
MViT_block godhj93/MobileViT/utils/nets/MobileViT.py community (archive-listed) ran no licence file found · pointer only · 0d8adae319bee6c7 · report
MobileViTAttention xmu-xiaoma666/External-Attention-pytorch/model/backbone/MobileViT.py community (archive-listed) ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 315d52a69ddead31 · report
MobileViTBlock jaiwei98/mobile-vit-pytorch/mobile_vit/mobilevit.py community (archive-listed) ran · metamorphic tier: deterministic no licence file found · pointer only · 51b0a869e04e4c79 · report
MobileViTBlock roydenwa/cell-centroid-former/src/cellcentroidformer/model_architecture.py community (archive-listed) ran no licence file found · pointer only · 6f15cb3df6e1d23e · report
MobileViTBlock murufeng/awesome_lightweight_networks/light_cnns/Transformer/mobile_vit.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 69a7b757b4410e76 · report
MobileViTBlock chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) ran MIT (permissive) · 8eca4842b31caba0 · report
MobileVitBlock wilile26811249/MobileViT/models/model.py community (archive-listed) ran MIT (permissive) · 355dd4ca53ea56d8 · report
MultiHeadSelfAttention wilile26811249/MobileViT/models/model.py community (archive-listed) ran MIT (permissive) · 3b57311302acf06c · report
MultiLayerPerceptron roydenwa/cell-centroid-former/src/cellcentroidformer/model_architecture.py community (archive-listed) ran no licence file found · pointer only · c6e2b5d57c742ded · report
PositionalEncodingFourier mmaaz60/EdgeNeXt/models/edgenext.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 9917b766bd4dba3a · report
PreNorm xmu-xiaoma666/External-Attention-pytorch/model/backbone/MobileViT.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · bcbab17e442a74cb · report
PreNorm kornia/kornia/kornia/models/vit_mobile.py community (archive-listed) ran · metamorphic tier: invariant Apache-2.0 (permissive) · 4bdddb9b8d76e1b6 · report
SDTAEncoder mmaaz60/EdgeNeXt/models/edgenext.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 7e438b1b13e38d9d · report
Transformer murufeng/awesome_lightweight_networks/light_cnns/Transformer/mobile_vit.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 127b3963437b49e3 · report
Transformer xmu-xiaoma666/External-Attention-pytorch/model/backbone/MobileViT.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 8ae59a8594d1df08 · report
Transformer wilile26811249/MobileViT/models/model.py community (archive-listed) ran MIT (permissive) · 54a5e3106186035e · report
Transformer chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) ran · metamorphic tier: invariant MIT (permissive) · 09bb5fbbbccebe6b · report
Transformer kornia/kornia/kornia/models/vit_mobile.py community (archive-listed) ran Apache-2.0 (permissive) · 56bdd39960d3695b · report
TransformerEncoder jaiwei98/mobile-vit-pytorch/mobile_vit/mobilevit.py community (archive-listed) ran · metamorphic tier: deterministic no licence file found · pointer only · bd36e71d0ee3fd10 · report
TransformerEncoder roydenwa/cell-centroid-former/src/cellcentroidformer/model_architecture.py community (archive-listed) ran no licence file found · pointer only · 50c31c18f8304002 · report
XCA mmaaz60/EdgeNeXt/models/edgenext.py community (archive-listed) ran · metamorphic tier: deterministic MIT (permissive) · 2733c35c71b87333 · report
compact_get_layers roydenwa/cell-centroid-former/src/cellcentroidformer/model_architecture.py community (archive-listed) ran · our draft was wrong no licence file found · pointer only · cf5208ed947ab445 · report
conv_1x1_bn chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 3fb34464dacbb0d3 · report
conv_1x1_bn kornia/kornia/kornia/models/vit_mobile.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · 564f22f3b28db627 · report
conv_bn xmu-xiaoma666/External-Attention-pytorch/model/backbone/MobileViT.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 3fc0fa5c567d1a28 · report
conv_nxn_bn chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) ran · our draft was wrong MIT (permissive) · e01bc963fa545868 · report
conv_nxn_bn kornia/kornia/kornia/models/vit_mobile.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · 8d5317047e54ef09 · report
count_parameters chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) ran · honoured contract MIT (permissive) · 3d1cb4db0a0a03e9 · report
extract_patches godhj93/MobileViT/utils/nets/MobileViT.py community (archive-listed) ran no licence file found · pointer only · f4179768dbc55e61 · report
patches_to_image godhj93/MobileViT/utils/nets/MobileViT.py community (archive-listed) ran no licence file found · pointer only · fa37ccabf8400bc6 · report
EdgeNeXt mmaaz60/EdgeNeXt/models/edgenext.py community (archive-listed) unverified MIT (permissive) · d8d7e7e34dc91a1e · report
InvertedResidual godhj93/MobileViT/utils/nets/MobileViT.py community (archive-listed) unverified no licence file found · pointer only · 6e05718496381a26 · report
MobileViT godhj93/MobileViT/utils/nets/MobileViT.py community (archive-listed) unverified no licence file found · pointer only · 86afc1196b6b0ee1 · report
MobileViT jaiwei98/mobile-vit-pytorch/mobile_vit/mobilevit.py community (archive-listed) unverified no licence file found · pointer only · 9758df99f0486f48 · report
MobileViT murufeng/awesome_lightweight_networks/light_cnns/Transformer/mobile_vit.py community (archive-listed) unverified MIT (permissive) · 773215bd26648b60 · report
MobileViT xmu-xiaoma666/External-Attention-pytorch/model/backbone/MobileViT.py community (archive-listed) unverified MIT (permissive) · aee9977341677a30 · report
MobileViT wilile26811249/MobileViT/models/model.py community (archive-listed) unverified MIT (permissive) · 36d81e4406f2c1da · report
MobileViT chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) unverified MIT (permissive) · d8f1a24b68aa0b29 · report
MobileViT kornia/kornia/kornia/models/vit_mobile.py community (archive-listed) unverified Apache-2.0 (permissive) · 1ac664c178170ffe · report
MobileViTBlock kornia/kornia/kornia/models/vit_mobile.py community (archive-listed) unverified Apache-2.0 (permissive) · 8be19ea313e7777e · report
MobileViT_S wilile26811249/MobileViT/models/model.py community (archive-listed) unverified MIT (permissive) · ed15506c976a883e · report
MobileViT_XS wilile26811249/MobileViT/models/model.py community (archive-listed) unverified MIT (permissive) · ccdc152785a294d2 · report
MobileViT_XXS wilile26811249/MobileViT/models/model.py community (archive-listed) unverified MIT (permissive) · 6014219e7ea97f79 · report
conv_1x1_bn chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) unverified MIT (permissive) · 248f0c82db73b84d · report
conv_nxn_bn chinhsuanwu/mobilevit-pytorch/mobilevit.py community (archive-listed) unverified MIT (permissive) · ba426aae4d6a9c16 · report
conv_1x1_bn identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · a0131fb70c267a9e · report
conv_2d identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · 46467acb70052484 · report

Tasks

Image ClassificationObject Detectionobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification ImageNet MobileViT-S Number of params 5.6M #837 of 1060 Archive leaderboard report
Image Classification ImageNet MobileViT-S Top 1 Accuracy 78.4% #837 of 1060 Archive leaderboard report
Image Classification ImageNet MobileViT-XS GFLOPs 0.7 #968 of 1060 Archive leaderboard report
Image Classification ImageNet MobileViT-XS Number of params 2.3M #968 of 1060 Archive leaderboard report
Image Classification ImageNet MobileViT-XS Top 1 Accuracy 74.8% #968 of 1060 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

Introduced by this paper: MobileViT

1x1 ConvolutionAttentionAttention DropoutAverage PoolingBatch NormalizationConvolutionDeiTDense ConnectionsDepthwise ConvolutionDepthwise Separable ConvolutionDropoutFeedforward NetworkGlobal Average PoolingHard SwishInverted Residual BlockLayer NormalizationLinear LayerMobileViTMulti-Head AttentionPointwise ConvolutionReLUReLU6Residual ConnectionSigmoid ActivationSoftmaxSqueeze-and-Excitation BlockVision Transformer

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