SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2206.04040 · 2022

MobileOne: An Improved One millisecond Mobile Backbone

arXiv · PDF · Open in the Atlas

Code that ran

4 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 1 failure of our instrument; 2 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/2206.04040.json · badge: /api/ran/2206.04040.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
apple/ml-mobileone canonical (classifier label)
classifier basis: classifier_2025
1 ran · 1 no recorded run
federicopozzi33/MobileOne-PyTorch community (classifier label)
classifier basis not recorded on the link
3 ran · 1 no recorded run
FunctionStatusWhere it lives
pad_1x1_to_3x3_tensor Ran
ran on a synthesized input
federicopozzi33/MobileOne-PyTorch/mobileone_pytorch/_reparametrize.py
code served (permissive licence) · get_code("8cb327644ca15c23")
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 Tensor, ndim 4, dtype family float, shape (2, 4, 3, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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.functional as F

torch.manual_seed(0)

# Create a small 1x1 kernel tensor
# Shape: (out_channels, in_channels, height, width) = (2, 4, 1, 1)
kernel1x1 = torch.randn(2, 4, 1, 1)

ARGS = {
    "kernel1x1": kernel1x1,
}
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 Tensor, ndim 4, dtype family float, shape (2, 4, 3, 3) — shape (2, 4, 3, 3) as a later run recorded it
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: none by the harness; any seed is in the setup block
rep_bn_to_conv2d Ran
ran on a synthesized input
federicopozzi33/MobileOne-PyTorch/mobileone_pytorch/_reparametrize.py
code served (permissive licence) · get_code("4ca57268f3fb57a1")
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 Conv2d. 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)

# Create a BatchNorm2d layer with num_features=4
batch_norm = nn.BatchNorm2d(num_features=4)

# Initialize the running statistics and parameters
batch_norm.running_mean = torch.randn(4)
batch_norm.running_var = torch.abs(torch.randn(4)) + 0.1
batch_norm.weight = nn.Parameter(torch.randn(4))
batch_norm.bias = nn.Parameter(torch.randn(4))
batch_norm.eps = 1e-5

# Test with kernel_size (1, 1)
kernel_size_1x1 = (1, 1)

# Test with kernel_size (3, 3)
kernel_size_3x3 = (3, 3)

# Define ARGS with kernel_size (3, 3) as primary test case
ARGS = {
    "batch_norm": batch_norm,
    "kernel_size": kernel_size_3x3,
    "groups": 1
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
rep_conv2d_bn_to_conv2d Ran
ran on a synthesized input
federicopozzi33/MobileOne-PyTorch/mobileone_pytorch/_reparametrize.py
code served (permissive licence) · get_code("dc9f04f8004c26ff")
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 Conv2d. 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 torch.nn.functional as F

torch.manual_seed(0)

# Create a Conv2d layer
in_channels = 3
out_channels = 4
kernel_size = (1, 1)
conv = nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, padding=(0, 0), bias=True)
conv.weight.data = torch.randn(out_channels, in_channels, 1, 1)
conv.bias.data = torch.randn(out_channels)

# Create a BatchNorm2d layer
batch_norm = nn.BatchNorm2d(out_channels)
batch_norm.running_mean = torch.randn(out_channels)
batch_norm.running_var = torch.abs(torch.randn(out_channels)) + 0.1
batch_norm.weight = nn.Parameter(torch.randn(out_channels))
batch_norm.bias = nn.Parameter(torch.randn(out_channels))
batch_norm.eps = 1e-5

ARGS = {
    "conv": conv,
    "batch_norm": batch_norm,
    "to_3x3": False
}
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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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
reparameterize_model Ran
ran; our contract draft was wrong, not the code
apple/ml-mobileone/mobileone.py
pointer only (licence: NOASSERTION) · get_code("26534874fe1cf631")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, apple/ml-mobileone (NOASSERTION). 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.

get_params Not yet run
unverified (harvested, no recorded run)
federicopozzi33/MobileOne-PyTorch/mobileone_pytorch/_mobileone_network.py
code served (permissive licence) · get_code("c0f962b136ec921d")
How it ran

no invocation recorded.

mobileone Not yet run
unverified (harvested, no recorded run)
apple/ml-mobileone/mobileone.py
pointer only (licence: NOASSERTION) · get_code("605c9a0436c7d190")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, apple/ml-mobileone (NOASSERTION). 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 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 2 later) and 4 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
reparameterize_model 3 other papers
1 dated earlier, 2 later
2203.03339 (2022-03)
L2CS-Net: Fine-Grained Gaze Estimation in Unconstrained Environments
2309.03452 (2023-09)
Multimodal Guidance Network for Missing-Modality Inference in Content Moderation
2406.05850 (2024-06)
Scaling Graph Convolutions for Mobile Vision
mobileone 1 other paper
1 later
2309.03452 (2023-09)
Multimodal Guidance Network for Missing-Modality Inference in Content Moderation

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

Efficient neural network backbones for mobile devices are often optimized for metrics such as FLOPs or parameter count. However, these metrics may not correlate well with latency of the network when deployed on a mobile device. Therefore, we perform extensive analysis of different metrics by deploying several mobile-friendly networks on a mobile device. We identify and analyze architectural and optimization bottlenecks in recent efficient neural networks and provide ways to mitigate these bottlenecks. To this end, we design an efficient backbone MobileOne, with variants achieving an inference time under 1 ms on an iPhone12 with 75.9% top-1 accuracy on ImageNet. We show that MobileOne achieves state-of-the-art performance within the efficient architectures while being many times faster on mobile. Our best model obtains similar performance on ImageNet as MobileFormer while being 38x faster. Our model obtains 2.3% better top-1 accuracy on ImageNet than EfficientNet at similar latency. Furthermore, we show that our model generalizes to multiple tasks - image classification, object detection, and semantic segmentation with significant improvements in latency and accuracy as compared to existing efficient architectures when deployed on a mobile device. Code and models are available at https://github.com/apple/ml-mobileone

For agents

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

get_harvested_code_for_paper("2206.04040")
get_code_for_paper("2206.04040")
have("2206.04040")

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

curl https://syntology.ai/api/ran/2206.04040.json

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

[![Syntology run record](https://syntology.ai/api/ran/2206.04040.svg)](https://syntology.ai/paper/2206.04040)

Connect an agent — have() is free.