Papers › DINOv2: Learning Robust Visual Features without Supervision

DINOv2: Learning Robust Visual Features without Supervision

14 Apr 2023arXiv:2304.07193archive 2025-07-28

Maxime Oquab, Timothée Darcet, Théo Moutakanni, Huy Vo, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, Mahmoud Assran, Nicolas Ballas, Wojciech Galuba, Russell Howes, Po-Yao Huang, Shang-Wen Li, Ishan Misra, Michael Rabbat, Vasu Sharma, Gabriel Synnaeve, Hu Xu, Hervé Jegou, Julien Mairal, Patrick Labatut, Armand Joulin, Piotr Bojanowski

The recent breakthroughs in natural language processing for model pretraining on large quantities of data have opened the way for similar foundation models in computer vision. These models could greatly simplify the use of images in any system by producing all-purpose visual features, i.e., features that work across image distributions and tasks without finetuning. This work shows that existing pretraining methods, especially self-supervised methods, can produce such features if trained on enough curated data from diverse sources. We revisit existing approaches and combine different techniques to scale our pretraining in terms of data and model size. Most of the technical contributions aim at accelerating and stabilizing the training at scale. In terms of data, we propose an automatic pipeline to build a dedicated, diverse, and curated image dataset instead of uncurated data, as typically done in the self-supervised literature. In terms of models, we train a ViT model (Dosovitskiy et al., 2020) with 1B parameters and distill it into a series of smaller models that surpass the best available all-purpose features, OpenCLIP (Ilharco et al., 2021) on most of the benchmarks at image and pixel levels.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

No sample from the archive's official repository (facebookresearch/dinov2) is linked to this paper; the 46 samples below come from 10 community repositories. Syntology harvested code from facebookresearch/dinov2 for another paper.

facebookresearch/dinov2: Syntology's graph links these samples to Vision Transformers Need Registers (arXiv:2309.16588; that paper's own run record: 15 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 1 honoured, 1 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method.

Syntology 46 code samples harvested from 10 repositories linked to this paper: 39 ran (of which 13 constructed an object rather than computing a result; 32 with no instrument failure: 2 honoured, 0 violated, 30 with no contract checked; 7 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; 5 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 46 samples from 10 repositories, 39 ran (32 with no instrument failure; 7 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 10 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 39 samples that ran has no recorded date.

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

facebookresearch/dinov2 officialmentioned in papermentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Vision Transformers Need Registers (arXiv:2309.16588; that paper's own run record: 15 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 1 honoured, 1 violated, 11 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. report
ByungKwanLee/Causal-Unsupervised-Segmentation mentioned on GitHubpytorchSyntology: 11 samples, 9 ran (2 instrument-failed) · 2 unverified report
JHKim-snu/PGA mentioned on GitHubpytorchSyntology: not harvested report
beneroth13/dinov2 mentioned on GitHubpytorchApache-2.0Syntology: 6 samples, 6 ran (4 instrument-failed) · 0 unverified report
bespontaneous/proteus-pytorch mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Accessing Vision Foundation Models at ImageNet-level Costs (arXiv:2407.10366; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
buyeah1109/KEN mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
buyeah1109/finc mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Unsupervised Learning of Visual Features by Contrasting Cluster Assignments (arXiv:2006.09882; that paper's own run record: 13 ran (of which 4 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 1 violated, 4 with no contract checked; 7 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report
fabio-sim/Depth-Anything-ONNX mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Depth Anything V2 (arXiv:2406.09414; that paper's own run record: 9 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 5 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
facebookresearch/highrescanopyheight mentioned on GitHubpytorchSyntology: 8 samples, 6 ran (0 instrument-failed) · 2 unverified report
gorkaydemir/DINOSAUR mentioned on GitHubpytorchSyntology: 5 samples, 4 ran (0 instrument-failed) · 1 unverified report
huggingface/transformers mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Attention Is All You Need (arXiv:1706.03762; that paper's own run record: 610 ran (of which 293 constructed an object rather than computing a result; 529 with no instrument failure: 45 honoured, 15 violated, 469 with no contract checked; 81 where Syntology's instrument failed) · 336 unverified), HuggingFace's Transformers: State-of-the-art Natural Language Processing (arXiv:1910.03771; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified), DeBERTa: Decoding-enhanced BERT with Disentangled Attention (arXiv:2006.03654; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 2 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 9 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
marrlab/dinobloom mentioned on GitHubpytorchApache-2.0Syntology: 2 samples, 2 ran (1 instrument-failed) · 0 unverified report
mohammedsb/dinov2formedical mentioned on GitHubpytorchNOASSERTIONSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Evaluating General Purpose Vision Foundation Models for Medical Image Analysis: An... (arXiv:2312.02366; that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 4 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
open-edge-platform/geti mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
roboflow/rf-detr mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Deformable DETR: Deformable Transformers for End-to-End Object Detection (arXiv:2010.04159; that paper's own run record: 37 ran (of which 9 constructed an object rather than computing a result; 24 with no instrument failure: 1 honoured, 3 violated, 20 with no contract checked; 13 where Syntology's instrument failed) · 18 unverified); nothing checks that they implement this paper's method. report
seatizendoi/dinovdeau mentioned on GitHubpytorchCC0-1.0Syntology: 10 samples, 10 ran (0 instrument-failed) · 0 unverified report
zhu-xlab/softcon mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
2024-MindSpore-1/Code2 mindsporeSyntology: not harvested report
BurguerJohn/global_perceptual_similarity_loss pytorchSyntology: 1 sample, 0 ran · 1 unverified report
BurguerJohn/torch-felix pytorchMITSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
OML-Team/open-metric-learning pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (arXiv:2010.11929; that paper's own run record: 307 ran (of which 165 constructed an object rather than computing a result; 286 with no instrument failure: 8 honoured, 2 violated, 276 with no contract checked; 21 where Syntology's instrument failed) · 112 unverified); nothing checks that they implement this paper's method. report
PaddlePaddle/PASSL paddleApache-2.0Syntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
leondgarse/keras_cv_attention_models tfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RepViT: Revisiting Mobile CNN From ViT Perspective (arXiv:2307.09283; that paper's own run record: 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 6 unverified); nothing checks that they implement this paper's method. report
open-edge-platform/training_extensions pytorchApache-2.0Syntology: not harvested report
pwc-1/Paper-8 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to MathOdyssey: Benchmarking Mathematical Problem-Solving Skills in Large Language Models... (arXiv:2406.18321; that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
gitlab.com/birder/birder pytorchSyntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

46 samples harvested: 39 ran (of which 13 constructed an object rather than computing a result; 32 with no instrument failure: 2 honoured, 0 violated, 30 with no contract checked; 7 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
2ran · our draft was wrong
5ran; the contract check could not run
30ran
7unverified

Licence: 12 of the 46 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 10 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 9 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

ByungKwanLee/Causal-Unsupervised-Segmentation community (archive-listed) 11 samples, 9 ran (2 instrument-failed) · 2 unverified
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
models/dinov2vit.py ran no licence file found · pointer only · 9557b19d4ebe95ac · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-24
Block in 252 papers
census: 5 implementations from 5 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
models/dinov2vit.py ran · metamorphic tier: deterministic no licence file found · pointer only · 0d03b714971c0501 · report fixture: returned Block · level recorded 2026-09-22
BlockChunk models/dinov2vit.py ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · 927d2139eab04c62 · report fixture: returned BlockChunk · level recorded 2026-09-22
DropPath in 75 papers
census: 17 implementations from 17 papers, 8 distinct outputs across 3 buckets, one shared input per bucket
models/dinov2vit.py ran · metamorphic tier: deterministic fingerprinted no licence file found · pointer only · 22e81d05c408426c · report fixture: returned DropPath · level recorded 2026-09-22
LayerScale in 15 papers models/dinov2vit.py ran · metamorphic tier: deterministic no licence file found · pointer only · 1d18daa479e53b33 · report fixture: returned LayerScale · level recorded 2026-09-22
Mlp in 102 papers models/dinov2vit.py ran · metamorphic tier: deterministic no licence file found · pointer only · 11ea568c90b9f65a · report fixture: returned Mlp · level recorded 2026-09-22
PatchEmbed in 123 papers
census: 9 implementations from 9 papers, 6 distinct outputs on one shared input
models/dinov2vit.py ran · metamorphic tier: invariant no licence file found · pointer only · 8b5fc434aefb49a7 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
drop_add_residual_stochastic_depth in 26 papers models/dinov2vit.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant no licence file found · pointer only · ba6d096d6c5fffc9 · report fixture: signature x:Tensor/3/float, residual_func:callable/0/?, sample_drop_ratio:float/0/float · returned Tensor · level recorded 2026-09-22
named_apply in 32 papers models/dinov2vit.py ran · our draft was wrong no licence file found · pointer only · 764b0f3c664386a8 · report fixture: signature fn:callable/0/?, module:callable/0/?, name:str/0/str, depth_first:bool/0/bool, include_root:bool/0/bool · returned SimpleModel · level recorded 2026-09-22
DinoVisionTransformer models/dinov2vit.py unverified no licence file found · pointer only · 64bfcff23fded249 · report
init_weights_vit_timm in 6 papers models/dinov2vit.py unverified no licence file found · pointer only · 9c2af68d5f7355a0 · report
beneroth13/dinov2 community (archive-listed) 6 samples, 6 ran (4 instrument-failed) · 0 unverified
add_residual in 18 papers dinov2/layers/block.py ran Apache-2.0 (permissive) · 5b78a828a7e415d5 · report fixture: signature x:Tensor/2/float, brange:Tensor/1/int, residual:Tensor/2/float, residual_scale_factor:float/0/float, scaling_vector:Tensor/2/float · returned Tensor · level recorded 2026-09-24
drop_add_residual_stochastic_depth in 26 papers dinov2/layers/block.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · 85f7ffc01945bb72 · report fixture: signature x:Tensor/3/float, residual_func:callable/0/?, sample_drop_ratio:float/0/float · returned Tensor · level recorded 2026-09-26
drop_path in 325 papers
census: 27 implementations from 61 papers, 5 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 4, kind float, dtype float32
dinov2/layers/drop_path.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant Apache-2.0 (permissive) · c157f5b112b3a392 · report fixture: signature x:Tensor/4/float, drop_prob:float/0/float, training:bool/0/bool · returned Tensor · level recorded 2026-09-22
get_branges_scales in 18 papers dinov2/layers/block.py ran; the contract check could not run (RAISES) fingerprinted Apache-2.0 (permissive) · 5be3610fa1fee19e · report fixture: signature x:Tensor/3/float, sample_drop_ratio:float/0/float · returned tuple · level recorded 2026-09-26
make_2tuple in 21 papers dinov2/layers/patch_embed.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 3c04490a215299f2 · report fixture: signature x:int/0/int · returned tuple · level recorded 2026-09-22
named_apply in 32 papers dinov2/models/vision_transformer.py ran · our draft was wrong Apache-2.0 (permissive) · e7fcb6d9ac9deaf4 · report fixture: signature fn:callable/0/?, module:callable/0/?, name:str/0/str, depth_first:bool/0/bool, include_root:bool/0/bool · returned SimpleModel · level recorded, date not recorded
buyeah1109/KEN community (archive-listed) 1 sample, 0 ran · 1 unverified
DINOv2FeatureExtractor KEN/features/DINOv2FeatureExtractor.py unverified no licence file found · pointer only · 2befc0a9ae12b450 · report
facebookresearch/highrescanopyheight community (archive-listed) 8 samples, 6 ran (0 instrument-failed) · 2 unverified
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket this sample: bucket 1 of 4, arg 1: rank 3, kind float, dtype float32
models/backbone.py ran · metamorphic tier: invariant Apache-2.0 (permissive) · 8598c6b96e3a0fbb · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-22
LayerScale in 15 papers models/backbone.py ran Apache-2.0 (permissive) · d502f73859574661 · report fixture: returned LayerScale · level recorded 2026-09-22
MaskingGenerator models/backbone.py ran Apache-2.0 (permissive) · 5fe0498944067403 · report fixture: returned MaskingGenerator · level recorded 2026-09-22
Mlp in 102 papers models/backbone.py ran Apache-2.0 (permissive) · a5b98a91fdfc68fa · report fixture: returned Mlp · level recorded 2026-09-22
PatchEmbed in 123 papers
census: 9 implementations from 9 papers, 6 distinct outputs on one shared input
models/backbone.py ran Apache-2.0 (permissive) · 421274bff5d2b1e9 · report fixture: returned PatchEmbed · level recorded 2026-09-22
make_2tuple in 21 papers models/backbone.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 1844e38d40beb412 · report fixture: signature x:int/0/int · returned tuple · level recorded 2026-09-22
DinoVisionTransformer models/backbone.py unverified Apache-2.0 (permissive) · d377a1c9f26035bb · report
gorkaydemir/DINOSAUR community (archive-listed) 5 samples, 4 ran (0 instrument-failed) · 1 unverified
Decoder in 210 papers models/model.py ran · metamorphic tier: deterministic MIT (permissive) · e9df43ed81dc2b08 · report fixture: returned Decoder · level recorded 2026-09-22
ISA models/model.py ran · metamorphic tier: deterministic MIT (permissive) · 05d603aa1a9a31eb · report fixture: returned ISA · level recorded 2026-09-22
MLP in 375 papers models/model.py ran · metamorphic tier: deterministic MIT (permissive) · 8912a968e28ea03d · report fixture: returned MLP · level recorded 2026-09-22
SA in 11 papers models/model.py ran · metamorphic tier: deterministic MIT (permissive) · 389e99f57cbc96c6 · report fixture: returned SA · level recorded 2026-09-22
DINOSAURpp models/model.py unverified MIT (permissive) · 165b58cc341915b6 · report
marrlab/dinobloom community (archive-listed) 2 samples, 2 ran (1 instrument-failed) · 0 unverified
add_residual in 18 papers dinov2/layers/block.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · 5d16d4d4fc573ac7 · report fixture: signature x:Tensor/2/float, brange:Tensor/1/int, residual:Tensor/2/float, residual_scale_factor:float/0/float, scaling_vector:NoneType/0/? · returned Tensor · level recorded 2026-09-26
drop_add_residual_stochastic_depth in 26 papers dinov2/layers/block.py ran Apache-2.0 (permissive) · 931b30d6ce9e37cb · report fixture: signature x:Tensor/3/float, residual_func:callable/0/?, sample_drop_ratio:float/0/float · returned Tensor · level recorded 2026-09-25
seatizendoi/dinovdeau community (archive-listed) 10 samples, 10 ran (0 instrument-failed) · 0 unverified
collate_fn_monolabel src/trainer/tools.py ran CC0-1.0 (permissive) · 707671e31e8906f9 · report fixture: signature batch:list/1/dict · returned dict · level recorded 2026-09-24
collate_fn_multilabel src/trainer/tools.py ran CC0-1.0 (permissive) · 7db18bed330a0eff · report fixture: signature batch:list/1/dict · returned dict · level recorded 2026-09-24
compute_best_threshold src/utils/evaluation.py ran fingerprinted CC0-1.0 (permissive) · 1dc3843893b5cd44 · report fixture: signature y_true:ndarray/1/int, y_probs:ndarray/1/float · returned tuple · level recorded 2026-09-24
compute_metrics_monolabel src/trainer/tools.py ran CC0-1.0 (permissive) · 2460ef2831016887 · report fixture: signature p:EvalPrediction/0/? · returned dict · level recorded 2026-09-24
create_head src/model/model_setup.py ran CC0-1.0 (permissive) · 4a4314d5371e5636 · report fixture: signature num_features:int/0/int, number_classes:int/0/int, dropout_prob:float/0/float, activation_func:callable/0/? · returned Sequential · level recorded 2026-09-24
format_transforms_to_markdown src/utils/model_card_generator.py ran CC0-1.0 (permissive) · d73ef6282cb6a0dc · report fixture: signature transforms_dict:dict/0/? · returned str · level recorded 2026-09-25
get_config_env src/utils/utils.py ran CC0-1.0 (permissive) · dfc557c31ab30db4 · report fixture: signature config_path:str/0/str · returned dict · level recorded 2026-09-24
get_training_type_from_args src/utils/enums.py ran CC0-1.0 (permissive) · 5e5ca5dc0d79533a · report fixture: signature args:Namespace/0/? · returned ClassificationType · level recorded 2026-09-24
logits_to_probs in 18 papers src/utils/evaluation.py ran fingerprinted CC0-1.0 (permissive) · 5474362c7aad954b · report fixture: signature predictions:ndarray/1/float · returned list · level recorded 2026-09-24
sigmoid in 216 papers
census: 36 implementations from 56 papers, 7 distinct outputs across 3 buckets, one shared input per bucket
src/utils/evaluation.py ran fingerprinted CC0-1.0 (permissive) · 66fcba2b19055cb7 · report fixture: signature _outputs:ndarray/2/float · returned ndarray · level recorded 2026-09-24
BurguerJohn/global_perceptual_similarity_loss community (archive-listed) 1 sample, 0 ran · 1 unverified
GlobalPercLossTeacherStudent GlobalPercLossTeacherStudent.py unverified MIT (permissive) · 316832e67ccf3416 · report
BurguerJohn/torch-felix community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
CreateWeights felix.py ran MIT (permissive) · 6ceb2adc39304f27 · report fixture: signature start_weight:float/0/float, end_weight:float/0/float, count:int/0/int, curve_weight:float/0/float · returned ndarray · level recorded 2026-09-25
PaddlePaddle/PASSL community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
wrap_forward passl/core/recompute.py ran Apache-2.0 (permissive) · c3b15e0d7daa5901 · report fixture: signature func:callable/0/?, newfunc:callable/0/? · returned function · level recorded 2026-09-25

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

4 assemblies on this paper: 4 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

4 sibling assemblies for DINOv2 (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for DINOv2 (graph method: DINOv2). 13 bricks; individually on a fixture: 11 ran (of which 5 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 4 siblings

  1. DinoVisionTransformer entry · level 0 · unverified · 64bfcff23fded249
  2. Block model · level 1 · ran · 0d03b714971c0501
  3. BlockChunk helper · level 1 · ran · 927d2139eab04c62
  4. Mlp model · level 1 · ran · 11ea568c90b9f65a
  5. PatchEmbed model · level 1 · ran · 8b5fc434aefb49a7
  6. init_weights_vit_timm helper · level 0 · unverified · 9c2af68d5f7355a0
  7. named_apply helper · level 1 · ran · our draft was wrong · 764b0f3c664386a8
  8. Attention model · level 1 · ran · 9557b19d4ebe95ac
  9. DropPath model · level 1 · ran · 22e81d05c408426c
  10. LayerScale model · level 1 · ran · 1d18daa479e53b33
  11. drop_add_residual_stochastic_depth helper · level 1 · ran; the contract check could not run (RAISES) · ba6d096d6c5fffc9
  12. drop_path helper · level 1 · ran; the contract check could not run (RAISES) · c157f5b112b3a392
  13. make_2tuple helper · level 1 · ran · honoured contract · 3c04490a215299f2

harvested Assembly for DINOv2 (graph method: DINOv2). 3 bricks; individually on a fixture: 2 ran (of which 2 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified; every one of the 2 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 4 siblings

  1. DINOv2FeatureExtractor entry · level 0 · unverified · 2befc0a9ae12b450
  2. ImageFeatureExtractor helper · level 1 · ran · d66b0649e0c37be9
  3. TransformedIndexDataset helper · level 1 · ran · ba4892e53d17ff07

harvested Assembly for DINOv2 (graph method: DINOv2). 8 bricks; individually on a fixture: 6 ran (of which 4 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 4 siblings

  1. DinoVisionTransformer entry · level 0 · unverified · d377a1c9f26035bb
  2. Block model · level 0 · unverified · 912e7d3558003565
  3. MaskingGenerator helper · level 1 · ran · 5fe0498944067403
  4. Mlp model · level 1 · ran · a5b98a91fdfc68fa
  5. PatchEmbed model · level 1 · ran · 421274bff5d2b1e9
  6. Attention model · level 1 · ran · 8598c6b96e3a0fbb
  7. LayerScale model · level 1 · ran · d502f73859574661
  8. make_2tuple helper · level 1 · ran · honoured contract · 1844e38d40beb412

harvested Assembly for DINOv2 (graph method: DINOv2). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

one of 4 siblings

  1. GlobalPercLossTeacherStudent entry · level 0 · unverified · 316832e67ccf3416

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2304.07193") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Depth EstimationDomain GeneralizationFine-Grained Image ClassificationImage ClassificationImage RetrievalMonocular Depth EstimationSelf-Supervised Image ClassificationSemantic SegmentationVisual Place Recognition

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Depth Estimation NYU-Depth V2 DINOv2 (ViT-g/14 frozen, w/ DPT decoder) RMS 0.279 #2 of 17 Archive leaderboard report
Domain Generalization ImageNet-C DINOv2 (ViT-g/14, frozen model, linear eval) Number of params 1100M #1 of 47 Archive leaderboard report
Domain Generalization ImageNet-C DINOv2 (ViT-g/14, frozen model, linear eval) mean Corruption Error (mCE) 28.2 #1 of 47 Archive leaderboard report
Domain Generalization ImageNet-C DINOv2 (ViT-L/14, frozen model, linear eval) Number of params 307M #4 of 47 Archive leaderboard report
Domain Generalization ImageNet-C DINOv2 (ViT-L/14, frozen model, linear eval) mean Corruption Error (mCE) 31.5 #4 of 47 Archive leaderboard report
Domain Generalization ImageNet-C DINOv2 (ViT-B/14, frozen model, linear eval) Number of params 85M #19 of 47 Archive leaderboard report
Domain Generalization ImageNet-C DINOv2 (ViT-B/14, frozen model, linear eval) mean Corruption Error (mCE) 42.7 #19 of 47 Archive leaderboard report
Domain Generalization ImageNet-C DINOv2 (ViT-S/14, frozen model, linear eval) Number of params 21M #31 of 47 Archive leaderboard report
Domain Generalization ImageNet-C DINOv2 (ViT-S/14, frozen model, linear eval) mean Corruption Error (mCE) 54.4 #31 of 47 Archive leaderboard report
Fine-Grained Image Classification Oxford-IIIT Pet Dataset DINOv2 (ViT-g/14, frozen model, linear eval) Accuracy 96.7 #3 of 15 Archive leaderboard report
Image Classification CIFAR-10 DINOv2 (ViT-g/14, frozen model, linear eval) Percentage correct 99.5 #2 of 265 Archive leaderboard report
Image Retrieval AmsterTime DINOv2 distilled (ViT-L/14 frozen) mAP 50.0 #1 of 5 Archive leaderboard report
Image Retrieval AmsterTime DINOv2 (ViT-g/14 frozen) mAP 46.7 #2 of 5 Archive leaderboard report
Image Retrieval AmsterTime DINOv2 distilled (ViT-B/14 frozen) mAP 45.6 #3 of 5 Archive leaderboard report
Image Retrieval AmsterTime DINOv2 distilled (ViT-S/14 frozen) mAP 43.5 #4 of 5 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split DINOv2 (ViT-g/14 frozen, w/ DPT decoder) Delta < 1.25 0.968 #40 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split DINOv2 (ViT-g/14 frozen, w/ DPT decoder) Delta < 1.25^2 0.997 #40 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split DINOv2 (ViT-g/14 frozen, w/ DPT decoder) Delta < 1.25^3 0.9993 #40 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split DINOv2 (ViT-g/14 frozen, w/ DPT decoder) RMSE 2.1128 #40 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split DINOv2 (ViT-g/14 frozen, w/ DPT decoder) RMSE log 0.0882 #40 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split DINOv2 (ViT-g/14 frozen, w/ DPT decoder) Sq Rel 0.1797 #40 of 79 Archive leaderboard report
Monocular Depth Estimation KITTI Eigen split DINOv2 (ViT-g/14 frozen, w/ DPT decoder) absolute relative error 0.0652 #40 of 79 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 DINOv2 (ViT-g/14 frozen, w/ DPT decoder) Delta < 1.25 0.9497 #40 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 DINOv2 (ViT-g/14 frozen, w/ DPT decoder) Delta < 1.25^2 0.996 #40 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 DINOv2 (ViT-g/14 frozen, w/ DPT decoder) Delta < 1.25^3 0.9994 #40 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 DINOv2 (ViT-g/14 frozen, w/ DPT decoder) RMSE 0.279 #40 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 DINOv2 (ViT-g/14 frozen, w/ DPT decoder) absolute relative error 0.0907 #40 of 85 Archive leaderboard report
Monocular Depth Estimation NYU-Depth V2 DINOv2 (ViT-g/14 frozen, w/ DPT decoder) log 10 0.0371 #40 of 85 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 (ViT-g/14 @448) Number of Params 1100M #2 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 (ViT-g/14 @448) Top 1 Accuracy 86.7% #2 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 (ViT-g/14) Number of Params 1100M #3 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 (ViT-g/14) Top 1 Accuracy 86.5% #3 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 distilled (ViT-L/14) Number of Params 307M #4 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 distilled (ViT-L/14) Top 1 Accuracy 86.3% #4 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 distilled (ViT-B/14) Number of Params 85M #7 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 distilled (ViT-B/14) Top 1 Accuracy 84.5% #7 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 distilled (ViT-S/14) Number of Params 21M #17 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet DINOv2 distilled (ViT-S/14) Top 1 Accuracy 81.1% #17 of 144 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) DINOv2 (ViT-g/14, 448) Number of Params 1100M #1 of 65 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) DINOv2 (ViT-g/14, 448) Top 1 Accuracy 88.9% #1 of 65 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) DINOv2 (ViT-g/14) Number of Params 1100M #3 of 65 Archive leaderboard report
Self-Supervised Image Classification ImageNet (finetuned) DINOv2 (ViT-g/14) Top 1 Accuracy 88.5% #3 of 65 Archive leaderboard report
Semantic Segmentation ADE20K DINOv2 (ViT-g/14 frozen model, w/ ViT-Adapter + Mask2former) Params (M) 1080 #13 of 235 Archive leaderboard report
Semantic Segmentation ADE20K DINOv2 (ViT-g/14 frozen model, w/ ViT-Adapter + Mask2former) Validation mIoU 60.2 #13 of 235 Archive leaderboard report
Semantic Segmentation Fine-Grained Grass Segmentation Dataset DINOv2 mIoU 47.57 #8 of 10 Archive leaderboard report
Visual Place Recognition 17 Places DINOv2 Recall@1 61.82 #5 of 8 Archive leaderboard report
Visual Place Recognition Baidu Mall DINOv2 Recall@1 49.21 #6 of 8 Archive leaderboard report
Visual Place Recognition Gardens Point DINOv2 Recall@1 71.50 #5 of 7 Archive leaderboard report
Visual Place Recognition Hawkins DINOv2 Recall@1 27.97 #6 of 7 Archive leaderboard report
Visual Place Recognition Laurel Caverns DINOv2 Recall@1 40.18 #3 of 7 Archive leaderboard report
Visual Place Recognition Mid-Atlantic Ridge DINOv2 Recall@1 24.75 #6 of 7 Archive leaderboard report
Visual Place Recognition Nardo-Air DINOv2 Recall@1 73.24 #2 of 7 Archive leaderboard report
Visual Place Recognition Nardo-Air R DINOv2 Recall@1 71.83 #6 of 8 Archive leaderboard report
Visual Place Recognition Oxford RobotCar Dataset DINOv2 Recall@1 39.79 #5 of 7 Archive leaderboard report
Visual Place Recognition Pittsburgh-30k-test DINOv2 Recall@1 78.32 #20 of 22 Archive leaderboard report
Visual Place Recognition St Lucia DINOv2 Recall@1 78.62 #10 of 14 Archive leaderboard report
Visual Place Recognition VP-Air DINOv2 Recall@1 45.23 #2 of 7 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

1-bit AdamAdam

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