Papers › RepVGG: Making VGG-style ConvNets Great Again

RepVGG: Making VGG-style ConvNets Great Again

11 Jan 2021CVPR 2021 1arXiv:2101.03697archive 2025-07-28

Xiaohan Ding, Xiangyu Zhang, Ningning Ma, Jungong Han, Guiguang Ding, Jian Sun

We present a simple but powerful architecture of convolutional neural network, which has a VGG-like inference-time body composed of nothing but a stack of 3x3 convolution and ReLU, while the training-time model has a multi-branch topology. Such decoupling of the training-time and inference-time architecture is realized by a structural re-parameterization technique so that the model is named RepVGG. On ImageNet, RepVGG reaches over 80% top-1 accuracy, which is the first time for a plain model, to the best of our knowledge. On NVIDIA 1080Ti GPU, RepVGG models run 83% faster than ResNet-50 or 101% faster than ResNet-101 with higher accuracy and show favorable accuracy-speed trade-off compared to the state-of-the-art models like EfficientNet and RegNet. The code and trained models are available at https://github.com/megvii-model/RepVGG.

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="2101.03697")

Code

Syntology Ran 13 of 16 code samples harvested from 8 repositories linked to this paper; 3 have no recorded run. Of those that ran: 4 ran · our draft was wrong; 1 ran · fixture could not drive it; 8 ran with no contract checked.

By repository: official repository: 2 samples from 1 repository, 2 ran; community (archive-listed): 12 samples from 7 repositories, 9 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.

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

DingXiaoH/RepVGG officialmentioned on GitHubpytorch report
Deci-AI/super-gradients mentioned on GitHubpytorch report
ZJCV/ZCls mentioned on GitHubpytorch report
benjaminjellis/Objax-RepVGG mentioned on GitHubpytorch report
frgfm/Holocron mentioned on GitHubpytorch report
imad08/Repvgg_pytorch mentioned on GitHubpytorch report
james77777778/keras-image-models mentioned on GitHubpytorchApache-2.0 report
lmk123568/RepVGG_Tutorial mentioned on GitHub report
rwightman/pytorch-image-models mentioned on GitHubpytorch report
upczww/TensorRT-RepVGG mentioned on GitHubpytorch report
xdhe1216/ACNet mentioned on GitHubtf report
PaddlePaddle/PaddleClas paddleApache-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

16 samples harvested; 13 ran; 0 honoured the contract we drafted; 3 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.

4ran · our draft was wrong
1ran · fixture could not drive it
8ran
3unverified

Licence: 6 of the 16 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 8 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.

RepVGGBlock DingXiaoH/RepVGG/repvgg.py official repository ran fingerprinted MIT (permissive) · ad028191534fda47 · report
SEBlock DingXiaoH/RepVGG/repvgg.py official repository ran MIT (permissive) · 7a720bff84e2e8d6 · report
RepBlock frgfm/Holocron/holocron/models/classification/repvgg.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · 8cbc55a0dcb68703 · report
RepBlock xmu-xiaoma666/External-Attention-pytorch/model/rep/repvgg.py community (archive-listed) ran MIT (permissive) · 6601973ab7354d1f · report
RepVGGBlock ZJCV/ZCls/zcls/model/layers/repvgg_block.py community (archive-listed) ran · metamorphic tier: deterministic fingerprinted Apache-2.0 (permissive) · 09ed5d0bcfc8fc7f · report
RepVGGBlock ShinyJay2/Deep_learning_Paper_implementations/RepVGG/model.py community (archive-listed) ran no licence file found · pointer only · fa3b384718cafe2b · report
RepVGGBlock Deci-AI/super-gradients/src/super_gradients/modules/repvgg_block.py community (archive-listed) ran fingerprinted Apache-2.0 (permissive) · bb892c13838b4194 · report
_conv_bn xmu-xiaoma666/External-Attention-pytorch/model/rep/repvgg.py community (archive-listed) ran · our draft was wrong MIT (permissive) · 236bac003e7e2b05 · report
block imad08/Repvgg_pytorch/block.py community (archive-listed) ran fingerprinted no licence file found · pointer only · a057a0ac99341954 · report
fcbn imad08/Repvgg_pytorch/block.py community (archive-listed) ran · our draft was wrong no licence file found · pointer only · 10656055c6c6e24b · report
fuse_conv_bn frgfm/Holocron/holocron/models/classification/repvgg.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · 3a9f6be6a5763710 · report
RepVGG frgfm/Holocron/holocron/models/classification/repvgg.py community (archive-listed) unverified Apache-2.0 (permissive) · 7e21a83ef0e30ffa · report
conv_sequence frgfm/Holocron/holocron/models/classification/repvgg.py community (archive-listed) unverified Apache-2.0 (permissive) · d2c1151642f901f4 · report
fusion_rep_vgg xdhe1216/ACNet/acnet_repvgg_dbb_block_utils.py community (archive-listed) unverified no licence file found · pointer only · 8a6587c3a5d8aaff · report
conv_bn identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · 7186655759897af6 · report
diff_model identical code first harvested elsewhere ran · fixture could not drive it fingerprinted licence of this copy not recorded · ee7045377b07d71d · report

Tasks

Image ClassificationSemantic Segmentation

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification ImageNet RepVGG-B2 GFLOPs 18.4 #807 of 1060 Archive leaderboard report
Image Classification ImageNet RepVGG-B2 Number of params 80.31M #807 of 1060 Archive leaderboard report
Image Classification ImageNet RepVGG-B2 Top 1 Accuracy 78.78% #807 of 1060 Archive leaderboard report
Image Classification ImageNet RepVGG-B2g4 GFLOPs 11.3 #829 of 1060 Archive leaderboard report
Image Classification ImageNet RepVGG-B2g4 Number of params 55.77M #829 of 1060 Archive leaderboard report
Image Classification ImageNet RepVGG-B2g4 Top 1 Accuracy 78.5% #829 of 1060 Archive leaderboard report
Semantic Segmentation Cityscapes val RepVGG-B2 mIoU 80.57% #49 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

Introduced by this paper: RepVGG

1x1 ConvolutionAverage PoolingBatch NormalizationConvolutionDense ConnectionsDepthwise ConvolutionDepthwise Separable ConvolutionDropoutEfficientNetGlobal Average PoolingInverted Residual BlockLinear LayerPointwise ConvolutionRMSPropReLURepVGGResidual ConnectionSGD with MomentumSigmoid ActivationSqueeze-and-Excitation BlockWeight Decay

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