Papers › EfficientViT: Multi-Scale Linear Attention for High-Resolution Dense Prediction

EfficientViT: Multi-Scale Linear Attention for High-Resolution Dense Prediction

29 May 2022arXiv:2205.14756archive 2025-07-28

Han Cai, Junyan Li, Muyan Hu, Chuang Gan, Song Han

High-resolution dense prediction enables many appealing real-world applications, such as computational photography, autonomous driving, etc. However, the vast computational cost makes deploying state-of-the-art high-resolution dense prediction models on hardware devices difficult. This work presents EfficientViT, a new family of high-resolution vision models with novel multi-scale linear attention. Unlike prior high-resolution dense prediction models that rely on heavy softmax attention, hardware-inefficient large-kernel convolution, or complicated topology structure to obtain good performances, our multi-scale linear attention achieves the global receptive field and multi-scale learning (two desirable features for high-resolution dense prediction) with only lightweight and hardware-efficient operations. As such, EfficientViT delivers remarkable performance gains over previous state-of-the-art models with significant speedup on diverse hardware platforms, including mobile CPU, edge GPU, and cloud GPU. Without performance loss on Cityscapes, our EfficientViT provides up to 13.9× and 6.2× GPU latency reduction over SegFormer and SegNeXt, respectively. For super-resolution, EfficientViT delivers up to 6.4x speedup over Restormer while providing 0.11dB gain in PSNR. For Segment Anything, EfficientViT delivers 48.9x higher throughput on A100 GPU while achieving slightly better zero-shot instance segmentation performance on COCO.

PaperPDFCodeCode 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="2205.14756")

Code

Syntology Ran 4 of 4 code samples harvested from 1 repository linked to this paper; 0 have no recorded run. Of those that ran: 1 ran · honoured contract; 1 ran · our draft was wrong; 2 ran · fixture could not drive it.

By repository: official repository: 4 samples from 1 repository, 4 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

mit-han-lab/efficientvit officialmentioned in papermentioned on GitHubpytorch report
rwightman/pytorch-image-models officialmentioned in papermentioned on GitHubpytorch 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

4 samples harvested; 4 ran; 1 honoured the contract we drafted; 0 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
1ran · our draft was wrong
2ran · fixture could not drive it

Licence: 0 of the 4 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 mit-han-lab/efficientvit. “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.

load_image mit-han-lab/efficientvit/applications/efficientvit_sam/demo_efficientvit_sam_model.py official repository ran · honoured contract Apache-2.0 (permissive) · bb3d4f05a88ee013 · report
accuracy mit-han-lab/efficientvit/applications/efficientvit_cls/eval_efficientvit_cls_model.py official repository ran · fixture could not drive it Apache-2.0 (permissive) · 35931102182b33e8 · report
cat_images mit-han-lab/efficientvit/applications/efficientvit_sam/demo_efficientvit_sam_model.py official repository ran · fixture could not drive it Apache-2.0 (permissive) · 799ae994e666d035 · report
draw_binary_mask mit-han-lab/efficientvit/applications/efficientvit_sam/demo_efficientvit_sam_model.py official repository ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 066e85f370a04fd7 · report

Tasks

Autonomous DrivingImage ClassificationImage SegmentationInstance SegmentationObject DetectionPredictionSemantic SegmentationSuper-ResolutionZero-Shot Instance Segmentationobject-detection

2 archive task tags without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification ImageNet EfficientViT-L2 (r384) GFLOPs 20 #181 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-L2 (r384) Number of params 64M #181 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-L2 (r384) Top 1 Accuracy 86% #181 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-L2 (r288) GFLOPs 11 #215 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-L2 (r288) Number of params 64M #215 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-L2 (r288) Top 1 Accuracy 85.6% #215 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-L1 (r224) GFLOPs 5.3 #310 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-L1 (r224) Number of params 53M #310 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-L1 (r224) Top 1 Accuracy 84.5% #310 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-B3 (r288) GFLOPs 6.5 #337 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-B3 (r288) Number of params 49M #337 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-B3 (r288) Top 1 Accuracy 84.2% #337 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-B3 (r224) GFLOPs 4 #420 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-B3 (r224) Top 1 Accuracy 83.5% #420 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-B2 (r256) GFLOPs 2.1 #511 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-B2 (r256) Number of params 24M #511 of 1060 Archive leaderboard report
Image Classification ImageNet EfficientViT-B2 (r256) Top 1 Accuracy 82.7% #511 of 1060 Archive leaderboard report
Semantic Segmentation ADE20K EfficientViT-B3 (r512) Validation mIoU 49 #142 of 235 Archive leaderboard report
Semantic Segmentation Cityscapes val EfficientViT-B3 (r1184x2368) mIoU 83.2 #27 of 99 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

1x1 ConvolutionAbsolute Position EncodingsAdamAttentionAverage PoolingBPEBatch NormalizationConvolutionDense ConnectionsDepthwise ConvolutionDepthwise Separable ConvolutionDropoutInverted Residual BlockLabel SmoothingLayer NormalizationLinear LayerMix-FFNMulti-Head AttentionPointwise ConvolutionPosition-Wise Feed-Forward LayerRMSPropReLUResidual ConnectionSegFormerSigmoid ActivationSoftmaxSqueeze-and-Excitation BlockTransformer

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