SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2101.03697 · CVPR · 2021

RepVGG: Making VGG-style ConvNets Great Again

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

arXiv · PDF · Open in the Atlas

Code that ran

13 of 16 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (7 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 5 failures of our instrument; 3 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2101.03697.json · badge: /api/ran/2101.03697.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
frgfm/Holocron community
Papers with Code archive
2 ran · 2 no recorded run
xmu-xiaoma666/External-Attention-pytorch community
Papers with Code archive
2 ran · 0 no recorded run
DingXiaoH/RepVGG marked official in Papers with Code
Papers with Code archive
2 ran · 0 no recorded run
imad08/Repvgg_pytorch community
Papers with Code archive
2 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
2 ran · 0 no recorded run
ZJCV/ZCls community
Papers with Code archive
1 ran · 0 no recorded run
ShinyJay2/Deep_learning_Paper_implementations community
Papers with Code archive
1 ran · 0 no recorded run
Deci-AI/super-gradients community
Papers with Code archive
1 ran · 0 no recorded run
xdhe1216/ACNet community
Papers with Code archive
0 ran · 1 no recorded run
FunctionStatusWhere it lives
RepBlock Ran
ran on a synthesized input · constructed an object; no result was computed
frgfm/Holocron/holocron/models/classification/repvgg.py
code served (permissive licence) · get_code("8cbc55a0dcb68703")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, RepBlock: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from typing import Callable

torch.manual_seed(0)

# Define a minimal norm_layer that returns nn.BatchNorm2d
def norm_layer_factory(num_channels: int) -> nn.Module:
    return nn.BatchNorm2d(num_channels)

# Define a minimal act_layer
act_layer = nn.ReLU(inplace=True)

# RepBlock parameters
inplanes = 4
planes = 4
stride = 1
identity = True

ARGS = {
    "inplanes": inplanes,
    "planes": planes,
    "stride": stride,
    "identity": identity,
    "act_layer": act_layer,
    "norm_layer": norm_layer_factory,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type RepBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a RepBlock output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
RepBlock Ran
ran on a synthesized input
xmu-xiaoma666/External-Attention-pytorch/model/rep/repvgg.py
code served (permissive licence) · get_code("6601973ab7354d1f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

INIT_ARGS = {
    'input_channel': 8,
    'output_channel': 12,
    'kernel_size': 3,
    'groups': 1,
    'stride': 1,
    'deploy': False,
    'use_se': False
}

ARGS = {
    'inputs': torch.randn(2, 8, 16, 16)
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
RepVGGBlock Ran
ran on a synthesized input · constructed an object; no result was computed
ZJCV/ZCls/zcls/model/layers/repvgg_block.py
code served (permissive licence) · get_code("09ed5d0bcfc8fc7f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, RepVGGBlock: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'in_channels': 3,
    'out_channels': 16,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
    'dilation': 1,
    'groups': 1,
    'padding_mode': 'zeros'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type RepVGGBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a RepVGGBlock output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
RepVGGBlock Ran
ran on a synthesized input · constructed an object; no result was computed
ShinyJay2/Deep_learning_Paper_implementations/RepVGG/model.py
pointer only (licence: NONE) · get_code("fa3b384718cafe2b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ShinyJay2/Deep_learning_Paper_implementations when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

RepVGGBlock Ran
ran on a synthesized input · constructed an object; no result was computed
Deci-AI/super-gradients/src/super_gradients/modules/repvgg_block.py
code served (permissive licence) · get_code("bb892c13838b4194")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, RepVGGBlock: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "in_channels": 4,
    "out_channels": 8,
    "stride": 1,
    "dilation": 1,
    "groups": 1,
    "activation_type": nn.ReLU,
    "activation_kwargs": {},
    "se_type": nn.Identity,
    "se_kwargs": {},
    "build_residual_branches": True,
    "use_residual_connection": True,
    "use_alpha": False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type RepVGGBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a RepVGGBlock output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
RepVGGBlock Ran
ran on a synthesized input · constructed an object; no result was computed
DingXiaoH/RepVGG/repvgg.py
code served (permissive licence) · get_code("ad028191534fda47")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, RepVGGBlock: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

# Construct a RepVGGBlock instance for testing
ARGS = {
    'in_channels': 3,
    'out_channels': 4,
    'kernel_size': 3,
    'stride': 1,
    'padding': 1,
    'dilation': 1,
    'groups': 1,
    'padding_mode': 'zeros',
    'deploy': False,
    'use_se': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type RepVGGBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a RepVGGBlock output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
SEBlock Ran
ran on a synthesized input · constructed an object; no result was computed
DingXiaoH/RepVGG/repvgg.py
code served (permissive licence) · get_code("7a720bff84e2e8d6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, SEBlock: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "input_channels": 16,
    "internal_neurons": 8,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SEBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a SEBlock output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
_conv_bn Ran
ran; our contract draft was wrong, not the code
xmu-xiaoma666/External-Attention-pytorch/model/rep/repvgg.py
code served (permissive licence) · get_code("236bac003e7e2b05")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Sequential. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'input_channel': 3,
    'output_channel': 64,
    'kernel_size': 3,
    'padding': 1,
    'stride': 1,
    'groups': 1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
block Ran
ran on a synthesized input · constructed an object; no result was computed
imad08/Repvgg_pytorch/block.py
pointer only (licence: NONE) · get_code("a057a0ac99341954")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, imad08/Repvgg_pytorch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

conv_bn Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from DingXiaoH/RepMLP
pointer only · get_code("7186655759897af6")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

diff_model Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
this paper's copy was not recorded; identical code first harvested from xdhe1216/ACNet
pointer only; the stored copy get_code serves, from xdhe1216/ACNet, is pointer only (licence: NONE) · get_code("ee7045377b07d71d")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

fcbn Ran
ran; our contract draft was wrong, not the code
imad08/Repvgg_pytorch/block.py
pointer only (licence: NONE) · get_code("10656055c6c6e24b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, imad08/Repvgg_pytorch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

fuse_conv_bn Ran
ran; our contract draft was wrong, not the code
frgfm/Holocron/holocron/models/classification/repvgg.py
code served (permissive licence) · get_code("3a9f6be6a5763710")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
from torch import nn

torch.manual_seed(0)

# Create a Conv2d layer with:
# - in_channels=3
# - out_channels=4
# - kernel_size=3
conv = nn.Conv2d(in_channels=3, out_channels=4, kernel_size=3, bias=False)

# Create a BatchNorm2d layer with:
# - num_features=4 (must match conv's out_channels)
bn = nn.BatchNorm2d(num_features=4)

# Initialize bn with some meaningful values for running_mean, running_var, weight, bias
bn.running_mean.fill_(0.1)
bn.running_var.fill_(0.5)
bn.weight.data.fill_(1.0)
bn.bias.data.fill_(0.2)
bn.eps = 1e-5

ARGS = {
    "conv": conv,
    "bn": bn,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
RepVGG Not yet run
unverified (harvested, no recorded run)
frgfm/Holocron/holocron/models/classification/repvgg.py
code served (permissive licence) · get_code("7e21a83ef0e30ffa")
How it ran

no invocation recorded.

conv_sequence Not yet run
unverified (harvested, no recorded run)
frgfm/Holocron/holocron/models/classification/repvgg.py
code served (permissive licence) · get_code("d2c1151642f901f4")
How it ran

no invocation recorded.

fusion_rep_vgg Not yet run
unverified (harvested, no recorded run)
xdhe1216/ACNet/acnet_repvgg_dbb_block_utils.py
pointer only (licence: NONE) · get_code("8a6587c3a5d8aaff")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, xdhe1216/ACNet when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 4 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 4 later) and 12 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
conv_bn
brick of this paper's assembly: helper
10 other papers
1 dated earlier, 9 later
2003.13678 (2020-03)
Designing Network Design Spaces
2105.01883 (2021-05)
RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition
2112.11081 (2021-12)
RepMLPNet: Hierarchical Vision MLP with Re-parameterized Locality
and 7 more not named here
conv_sequence
brick of this paper's assembly: helper
2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
1505.04597 (2015-05)
U-Net: Convolutional Networks for Biomedical Image Segmentation
2201.03545 (2022-01)
A ConvNet for the 2020s
diff_model
brick of this paper's assembly: helper
2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
1908.03930 (2019-08)
ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks
2103.13425 (2021-03)
Diverse Branch Block: Building a Convolution as an Inception-like Unit
fusion_rep_vgg
brick of this paper's assembly: entry
1 other paper
1 later · 1 only as an assembly brick
2103.13425 (2021-03)
Diverse Branch Block: Building a Convolution as an Inception-like Unit

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We present a simple but powerful architecture of convolutional neural network, which has a VGG-like inferencetime body composed of nothing but a stack of 3 × 3 convolution and ReLU, while the training-time model has a multi-branch topology. Such decoupling of the trainingtime 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 stateof-the-art models like EfficientNet and RegNet. The code and trained models are available at https://github. com/megvii-model/RepVGG.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2101.03697")
get_code_for_paper("2101.03697")
have("2101.03697")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2101.03697.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2101.03697.svg)](https://syntology.ai/paper/2101.03697)

Connect an agent — have() is free.