Browse › Computer Vision › Semantic Segmentation › ADE20K
Semantic Segmentation archive 2025-07-28
ADE20K Benchmark (Semantic Segmentation)
Over time archive 2025-07-28
The chart needs JavaScript; the table below carries every value.
Direction inferred from the metric name, not from the archive: Test Score (higher is better), Params (M) (lower is better), Mean IoU (class) (higher is better). Not inferred (points only, no best-so-far line): Validation mIoU, GFLOPs (512 x 512), GFLOPs. Points are placed at the row's paper date; 235 of 235 rows carry one.
Results archive 2025-07-28
Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.
| Paper | Code | Ran Syntology | Report | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | ViT-P (InternImage-H) | 63.6 | 1610 | ✓ | Paper | Code | 2025 | linked, not harvested | report | ||||
| 2 | ONE-PEACE | 63.0 | 1500 | ✓ | Paper | Code | 2023 | 2 of 7 ran · 5 unverified | report | ||||
| 3 | InternImage-H | 62.9 | 1310 | 4635 | ✓ | Paper | Code | 2022 | 2 of 4 ran · 2 unverified | report | |||
| 4 | M3I Pre-training (InternImage-H) | 62.9 | 1310 | ✓ | Paper | Code | 2022 | linked, not harvested | report | ||||
| 5 | BEiT-3 | 62.8 | 1900 | ✓ | Paper | Code | 2022 | linked, not harvested | report | ||||
| 6 | EVA | 62.3 | 1074 | ✓ | Paper | Code | 2022 | 1 of 3 ran · 2 unverified | report | ||||
| 7 | ViT-P (OneFormer, InternImage-H) | 61.6 | 1400 | – | Paper | Code | 2025 | linked, not harvested | report | ||||
| 8 | ViT-Adapter-L (Mask2Former, BEiTv2 pretrain) | 61.5 | 571 | ✓ | Paper | Code | 2022 | linked, not harvested | report | ||||
| 9 | FD-SwinV2-G | 61.4 | 3000 | – | Paper | Code | 2022 | 2 of 8 ran · 6 unverified | report | ||||
| 10 | RevCol-H (Mask2Former) | 61.0 | 2439 | ✓ | Paper | Code | 2022 | 8 of 11 ran · 3 unverified | report | ||||
| 11 | MasK DINO (SwinL, multi-scale) | 60.8 | 223 | ✓ | Paper | Code | 2022 | 11 of 13 ran · 2 unverified | report | ||||
| 12 | ViT-Adapter-L (Mask2Former, BEiT pretrain) | 60.5 | 571 | ✓ | Paper | Code | 2022 | linked, not harvested | report | ||||
| 13 | DINOv2 (ViT-g/14 frozen model, w/ ViT-Adapter + Mask2former) | 60.2 | 1080 | – | Paper | Code | 2023 | 21 of 46 ran · 25 unverified | report | ||||
| 14 | ViT-P (OneFormer, DiNAT-L) | 59.9 | 309 | – | Paper | Code | 2025 | linked, not harvested | report | ||||
| 15 | SwinV2-G(UperNet) | 59.9 | ✓ | Paper | Code | 2021 | 3 of 30 ran · 27 unverified | report | |||||
| 16 | PIIP-LH6B(UperNet) | 59.9 | – | Paper | Code | 2024 | linked, not harvested | report | |||||
| 17 | SERNet-Former | 59.35 | – | Paper | Code | 2024 | linked, not harvested | report | |||||
| 18 | FocalNet-L (Mask2Former) | 58.5 | ✓ | Paper | Code | 2022 | 0 of 2 ran · 2 unverified | report | |||||
| 19 | ViT-Adapter-L (UperNet, BEiT pretrain) | 58.4 | 451 | – | Paper | Code | 2022 | linked, not harvested | report | ||||
| 20 | RSSeg-ViT-L (BEiT pretrain) | 58.4 | 330 | – | Paper | – | 2022 | no code linked | report | ||||
| 21 | EoMT (DINOv2-L, single-scale, 512x512) | 58.4 | 316 | 721 | 721 | 58.4 | – | Paper | Code | 2025 | 1 of 2 ran · 1 unverified | report | |
| 22 | SegViT-v2 (BEiT-v2-Large) | 58.2 | 637.9 | – | Paper | Code | 2023 | linked, not harvested | report | ||||
| 23 | SeMask (SeMask Swin-L FaPN-Mask2Former) | 58.2 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 24 | SeMask (SeMask Swin-L MSFaPN-Mask2Former) | 58.2 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 25 | DiNAT-L (Mask2Former) | 58.1 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 26 | HorNet-L (Mask2Former) | 57.9 | – | Paper | Code | 2022 | 1 of 1 ran · 0 unverified | report | |||||
| 27 | Mask2Former (SwinL-FaPN) | 57.7 | – | Paper | Code | 2021 | 2 of 8 ran · 6 unverified | report | |||||
| 28 | FASeg (SwinL) | 57.7 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 29 | RR (BEiT-L) | 57.7 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 30 | MOAT-4 (IN-22K pretraining, single-scale) | 57.6 | 496 | – | Paper | Code | 2022 | linked, not harvested | report | ||||
| 31 | Frozen Backbone, SwinV2-G-ext22K (Mask2Former) | 57.6 | – | Paper | – | 2022 | no code linked | report | |||||
| 32 | SeMask (SeMask Swin-L Mask2Former) | 57.5 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 33 | Mask2Former (SwinL) | 57.3 | – | Paper | Code | 2021 | 2 of 8 ran · 6 unverified | report | |||||
| 34 | SenFormer (BEiT-L) | 57.1 | ✓ | Paper | Code | 2021 | linked, not harvested | report | |||||
| 35 | BEiT-L (ViT+UperNet) | 57.0 | – | Paper | Code | 2021 | 6 of 11 ran · 5 unverified | report | |||||
| 36 | SeMask(SeMask Swin-L MSFaPN-Mask2Former, single-scale) | 57.0 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 37 | MetaPrompt-SD | 56.8 | – | Paper | Code | 2023 | 4 of 6 ran · 2 unverified | report | |||||
| 38 | FaPN (MaskFormer, Swin-L, ImageNet-22k pretrain) | 56.7 | – | Paper | Code | 2021 | 2 of 4 ran · 2 unverified | report | |||||
| 39 | MOAT-3 (IN-22K pretraining, single-scale) | 56.5 | 198 | – | Paper | Code | 2022 | linked, not harvested | report | ||||
| 40 | Mask2Former (Swin-L-FaPN) | 56.4 | – | Paper | Code | 2021 | 2 of 8 ran · 6 unverified | report | |||||
| 41 | SeMask (SeMask Swin-L MaskFormer) | 56.2 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 42 | dBOT ViT-L (CLIP) | 56.2 | – | Paper | Code | 2022 | 5 of 13 ran · 8 unverified | report | |||||
| 43 | Mask2Former+CBL(Swin-B) | 56.1 | – | Paper | Code | 2023 | linked, not harvested | report | |||||
| 44 | TADP | 55.9 | – | Paper | Code | 2023 | linked, not harvested | report | |||||
| 45 | CSWin-L (UperNet, ImageNet-22k pretrain) | 55.70 | – | Paper | Code | 2021 | 2 of 6 ran · 4 unverified | report | |||||
| 46 | UniRepLKNet-XL | 55.6 | – | Paper | Code | 2023 | 1 of 2 ran · 1 unverified | report | |||||
| 47 | Focal-L (UperNet, ImageNet-22k pretrain) | 55.40 | – | Paper | Code | 2021 | 3 of 3 ran · 0 unverified | report | |||||
| 48 | InternImage-XL | 55.3 | 368 | 3142 | – | Paper | Code | 2022 | 2 of 4 ran · 2 unverified | report | |||
| 49 | dBOT ViT-L | 55.2 | – | Paper | Code | 2022 | 5 of 13 ran · 8 unverified | report | |||||
| 50 | Mask2Former(Swin-B) | 55.1 | – | Paper | Code | 2021 | 2 of 8 ran · 6 unverified | report | |||||
| 51 | ConvNeXt V2-H (FCMAE) | 55 | – | Paper | Code | 2023 | 10 of 14 ran · 4 unverified | report | |||||
| 52 | UniRepLKNet-L++ | 55 | – | Paper | Code | 2023 | 1 of 2 ran · 1 unverified | report | |||||
| 53 | DiNAT-Large (UperNet) | 54.9 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 54 | MaskFormer+CBL(Swin-B) | 54.9 | – | Paper | Code | 2023 | linked, not harvested | report | |||||
| 55 | TransNeXt-Base (IN-1K pretrain, Mask2Former, 512) | 54.7 | 109 | – | Paper | Code | 2023 | 5 of 5 ran · 0 unverified | report | ||||
| 56 | MOAT-2 (IN-22K pretraining, single-scale) | 54.7 | 81 | – | Paper | Code | 2022 | linked, not harvested | report | ||||
| 57 | CAE (ViT-L, UperNet) | 54.7 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 58 | VAN-B6 | 54.7 | – | Paper | Code | 2022 | 0 of 6 ran · 6 unverified | report | |||||
| 59 | DiNAT_s-Large (UperNet) | 54.6 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 60 | DDP (Swin-L, step-3) | 54.4 | 207 | – | Paper | Code | 2023 | linked, not harvested | report | ||||
| 61 | PatchDiverse + Swin-L (multi-scale test, upernet, ImageNet22k pretrain) | 54.4 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 62 | VOLO-D5 | 54.3 | – | Paper | Code | 2021 | 1 of 6 ran · 5 unverified | report | |||||
| 63 | K-Net | 54.3 | – | Paper | Code | 2021 | 0 of 1 ran · 1 unverified | report | |||||
| 64 | GPaCo (Swin-L) | 54.3 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 65 | SenFormer (Swin-L) | 54.2 | ✓ | Paper | Code | 2021 | linked, not harvested | report | |||||
| 66 | Swin V2-H | 54.2 | – | Paper | Code | 2023 | 10 of 14 ran · 4 unverified | report | |||||
| 67 | InternImage-L | 54.1 | 256 | 2526 | – | Paper | Code | 2022 | 2 of 4 ran · 2 unverified | report | |||
| 68 | TransNeXt-Small (IN-1K pretrain, Mask2Former, 512) | 54.1 | 69 | – | Paper | Code | 2023 | 5 of 5 ran · 0 unverified | report | ||||
| 69 | ConvNeXt-XL++ | 54 | 391 | 3335 | – | Paper | Code | 2022 | 54 of 80 ran · 26 unverified | report | |||
| 70 | Sequential Ensemble (SegFormer) | 54 | 216.3 | – | Paper | – | 2022 | no code linked | report | ||||
| 71 | MogaNet-XL (UperNet) | 54 | – | Paper | Code | 2022 | 12 of 15 ran · 3 unverified | report | |||||
| 72 | UniRepLKNet-B++ | 53.9 | – | Paper | Code | 2023 | 1 of 2 ran · 1 unverified | report | |||||
| 73 | MaskFormer(Swin-B) | 53.8 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 74 | ConvNeXt-L++ | 53.7 | 235 | 2458 | – | Paper | Code | 2022 | 54 of 80 ran · 26 unverified | report | |||
| 75 | SwinV2-G-HTC++ Liu et al. ([2021a]) | 53.7 | – | Paper | Code | 2021 | 3 of 30 ran · 27 unverified | report | |||||
| 76 | ConvNeXt V2-L | 53.7 | – | Paper | Code | 2023 | 10 of 14 ran · 4 unverified | report | |||||
| 77 | Seg-L-Mask/16 (MS) | 53.63 | – | Paper | Code | 2021 | 0 of 7 ran · 7 unverified | report | |||||
| 78 | MAE (ViT-L, UperNet) | 53.6 | – | Paper | Code | 2021 | 71 of 137 ran · 66 unverified | report | |||||
| 79 | SeMask (SeMask Swin-L FPN) | 53.52 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 80 | Swin-L (UperNet, ImageNet-22k pretrain) | 53.50 | 62.8 | – | Paper | Code | 2021 | 108 of 207 ran · 99 unverified | report | ||||
| 81 | Swin-L | 53.5 | – | Paper | Code | 2023 | 10 of 14 ran · 4 unverified | report | |||||
| 82 | TransNeXt-Tiny (IN-1K pretrain, Mask2Former, 512) | 53.4 | 47.5 | – | Paper | Code | 2023 | 5 of 5 ran · 0 unverified | report | ||||
| 83 | ConvNeXt-B++ | 53.1 | 122 | 1828 | – | Paper | Code | 2022 | 54 of 80 ran · 26 unverified | report | |||
| 84 | PatchConvNet-L120 (UperNet) | 52.9 | – | Paper | Code | 2021 | 1 of 2 ran · 1 unverified | report | |||||
| 85 | dBOT ViT-B (CLIP) | 52.9 | – | Paper | Code | 2022 | 5 of 13 ran · 8 unverified | report | |||||
| 86 | PatchConvNet-B120 (UperNet) | 52.8 | – | Paper | Code | 2021 | 1 of 2 ran · 1 unverified | report | |||||
| 87 | Swin-B | 52.8 | – | Paper | Code | 2023 | 10 of 14 ran · 4 unverified | report | |||||
| 88 | UniRepLKNet-S++ | 52.7 | – | Paper | Code | 2023 | 1 of 2 ran · 1 unverified | report | |||||
| 89 | ConvNeXt V2-B | 52.1 | – | Paper | Code | 2023 | 10 of 14 ran · 4 unverified | report | |||||
| 90 | DeBiFormer-B (IN1k pretrain, Upernet 160k) | 52.0 | – | Paper | Code | 2024 | linked, not harvested | report | |||||
| 91 | LV-ViT-L (UperNet, MS) | 51.8 | 209 | – | Paper | Code | 2021 | 3 of 5 ran · 2 unverified | report | ||||
| 92 | SegFormer-B5 | 51.8 | 84.7 | ✓ | Paper | Code | 2021 | 48 of 86 ran · 38 unverified | report | ||||
| 93 | BiFormer-B (IN1k pretrain, Upernet 160k) | 51.7 | – | Paper | Code | 2023 | 6 of 6 ran · 0 unverified | report | |||||
| 94 | ConvNeXt V2-L (Supervised) | 51.6 | – | Paper | Code | 2023 | 10 of 14 ran · 4 unverified | report | |||||
| 95 | Light-Ham (VAN-Huge) | 51.5 | 61.1 | 71.8 | – | Paper | Code | 2021 | 3 of 3 ran · 0 unverified | report | |||
| 96 | DAT-B++ | 51.5 | – | Paper | Code | 2023 | 2 of 10 ran · 8 unverified | report | |||||
| 97 | CrossFormer (ImageNet1k-pretrain, UPerNet, multi-scale test) | 51.4 | – | Paper | Code | 2021 | 1 of 1 ran · 0 unverified | report | |||||
| 98 | InternImage-B | 51.3 | 128 | 1185 | – | Paper | Code | 2022 | 2 of 4 ran · 2 unverified | report | |||
| 99 | DAT-S++ | 51.2 | – | Paper | Code | 2023 | 2 of 10 ran · 8 unverified | report | |||||
| 100 | ActiveMLP-L(UperNet) | 51.1 | 108 | – | Paper | Code | 2022 | linked, not harvested | report | ||||
| 101 | SegFormer-B4 | 51.1 | 64.1 | ✓ | Paper | Code | 2021 | 48 of 86 ran · 38 unverified | report | ||||
| 102 | PatchConvNet-B60 (UperNet) | 51.1 | – | Paper | Code | 2021 | 1 of 2 ran · 1 unverified | report | |||||
| 103 | Light-Ham (VAN-Large) | 51.0 | 45.6 | 55.0 | – | Paper | Code | 2021 | 3 of 3 ran · 0 unverified | report | |||
| 104 | TEC (Vit-B, Upernet) | 51.0 | – | Paper | Code | 2022 | 6 of 8 ran · 2 unverified | report | |||||
| 105 | UniRepLKNet-S | 51 | – | Paper | Code | 2023 | 1 of 2 ran · 1 unverified | report | |||||
| 106 | SeMask (SeMask Swin-B FPN) | 50.98 | 96 | – | Paper | Code | 2021 | linked, not harvested | report | ||||
| 107 | InternImage-S | 50.9 | 80 | 1017 | – | Paper | Code | 2022 | 2 of 4 ran · 2 unverified | report | |||
| 108 | MogaNet-L (UperNet) | 50.9 | 1176 | – | Paper | Code | 2022 | 12 of 15 ran · 3 unverified | report | ||||
| 109 | dBOT ViT-B | 50.8 | – | Paper | Code | 2022 | 5 of 13 ran · 8 unverified | report | |||||
| 110 | Upernet-BiFormer-S (IN1k pretrain, Upernet 160k) | 50.8 | – | Paper | Code | 2023 | 6 of 6 ran · 0 unverified | report | |||||
| 111 | UperNet Shuffle-B | 50.5 | – | Paper | Code | 2021 | 5 of 12 ran · 7 unverified | report | |||||
| 112 | ConvNeXt V1-L | 50.5 | – | Paper | Code | 2023 | 10 of 14 ran · 4 unverified | report | |||||
| 113 | DiNAT-Base (UperNet) | 50.4 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 114 | ELSA-Swin-S | 50.3 | – | Paper | Code | 2021 | 1 of 1 ran · 0 unverified | report | |||||
| 115 | DAT-T++ | 50.3 | – | Paper | Code | 2023 | 2 of 10 ran · 8 unverified | report | |||||
| 116 | SETR-MLA (160k, MS) | 50.28 | – | Paper | Code | 2020 | linked, not harvested | report | |||||
| 117 | VAN-Large (HamNet) | 50.2 | 55 | – | Paper | Code | 2022 | 0 of 6 ran · 6 unverified | report | ||||
| 118 | HRViT-b3 (SegFormer, SS) | 50.2 | 28.7 | 67.9 | – | Paper | Code | 2021 | linked, not harvested | report | |||
| 119 | Twins-SVT-L (UperNet, ImageNet-1k pretrain) | 50.2 | – | Paper | Code | 2021 | 0 of 2 ran · 2 unverified | report | |||||
| 120 | MogaNet-B (UperNet) | 50.1 | 1050 | – | Paper | Code | 2022 | 12 of 15 ran · 3 unverified | report | ||||
| 121 | Seg-B-Mask/16(MS, ViT-B) | 50.0 | – | Paper | Code | 2021 | 0 of 7 ran · 7 unverified | report | |||||
| 122 | iBOT (ViT-B/16) | 50.0 | – | Paper | Code | 2021 | 0 of 1 ran · 1 unverified | report | |||||
| 123 | ConvNeXt-B | 49.9 | 122 | 1170 | – | Paper | Code | 2022 | 54 of 80 ran · 26 unverified | report | |||
| 124 | DiNAT-Small (UperNet) | 49.9 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 125 | ConvNeXt V1-B | 49.9 | – | Paper | Code | 2023 | 10 of 14 ran · 4 unverified | report | |||||
| 126 | NAT-Base | 49.7 | 123 | 1137 | – | Paper | Code | 2022 | 1 of 1 ran · 0 unverified | report | |||
| 127 | Swin-B (UperNet, ImageNet-1k pretrain) | 49.7 | – | Paper | Code | 2021 | 108 of 207 ran · 99 unverified | report | |||||
| 128 | Seg-B/8 (MS, ViT-B) | 49.61 | – | Paper | Code | 2021 | 0 of 7 ran · 7 unverified | report | |||||
| 129 | ConvNeXt-S | 49.6 | 82 | 1027 | – | Paper | Code | 2022 | 54 of 80 ran · 26 unverified | report | |||
| 130 | Light-Ham (VAN-Base) | 49.6 | 27.4 | 34.4 | – | Paper | Code | 2021 | 3 of 3 ran · 0 unverified | report | |||
| 131 | NAT-Small | 49.5 | 82 | 1010 | – | Paper | Code | 2022 | 1 of 1 ran · 0 unverified | report | |||
| 132 | DaViT-B | 49.4 | – | Paper | Code | 2022 | 8 of 15 ran · 7 unverified | report | |||||
| 133 | DAT-B (UperNet) | 49.38 | 121 | – | Paper | Code | 2022 | 8 of 10 ran · 2 unverified | report | ||||
| 134 | PatchConvNet-S60 (UperNet) | 49.3 | – | Paper | Code | 2021 | 1 of 2 ran · 1 unverified | report | |||||
| 135 | ColorMAE-Green-ViTB-1600 | 49.3 | – | Paper | Code | 2024 | linked, not harvested | report | |||||
| 136 | MogaNet-S (UperNet) | 49.2 | 946 | – | Paper | Code | 2022 | 12 of 15 ran · 3 unverified | report | ||||
| 137 | Shift-B (UperNet) | 49.2 | – | Paper | Code | 2022 | 6 of 7 ran · 1 unverified | report | |||||
| 138 | UniRepLKNet-T | 49.1 | – | Paper | Code | 2023 | 1 of 2 ran · 1 unverified | report | |||||
| 139 | DPT-Hybrid | 49.02 | – | Paper | Code | 2021 | 54 of 116 ran · 62 unverified | report | |||||
| 140 | GC ViT-B | 49 | 125 | 1348 | – | Paper | Code | 2022 | 17 of 36 ran · 19 unverified | report | |||
| 141 | A2MIM (ViT-B) | 49 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 142 | EfficientViT-B3 (r512) | 49 | – | Paper | Code | 2022 | 4 of 4 ran · 0 unverified | report | |||||
| 143 | DiNAT-Tiny (UperNet) | 48.8 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 144 | HRViT-b2 (SegFormer, SS) | 48.76 | 20.8 | 28.0 | – | Paper | Code | 2021 | linked, not harvested | report | |||
| 145 | NAT-Tiny | 48.4 | 58 | 934 | – | Paper | Code | 2022 | 1 of 1 ran · 0 unverified | report | |||
| 146 | XCiT-M24/8 (UperNet) | 48.4 | – | Paper | Code | 2021 | 3 of 14 ran · 11 unverified | report | |||||
| 147 | ResNeSt-200 | 48.36 | – | Paper | Code | 2020 | 8 of 48 ran · 40 unverified | report | |||||
| 148 | DAT-S (UperNet) | 48.31 | 81 | – | Paper | Code | 2022 | 8 of 10 ran · 2 unverified | report | ||||
| 149 | GC ViT-S | 48.3 | 84 | 1163 | – | Paper | Code | 2022 | 17 of 36 ran · 19 unverified | report | |||
| 150 | InternImage-T | 48.1 | 59 | 944 | – | Paper | Code | 2022 | 2 of 4 ran · 2 unverified | report | |||
| 151 | VAN-Large | 48.1 | 49 | – | Paper | Code | 2022 | 0 of 6 ran · 6 unverified | report | ||||
| 152 | XCiT-S24/8 (UperNet) | 48.1 | – | Paper | Code | 2021 | 3 of 14 ran · 11 unverified | report | |||||
| 153 | MaskFormer(ResNet-101) | 48.1 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 154 | MAE (ViT-B, UperNet) | 48.1 | – | Paper | Code | 2021 | 71 of 137 ran · 66 unverified | report | |||||
| 155 | HRNetV2 + OCR + RMI (PaddleClas pretrained) | 47.98 | – | Paper | Code | 2019 | 4 of 9 ran · 5 unverified | report | |||||
| 156 | Shift-B | 47.9 | – | Paper | Code | 2022 | 6 of 7 ran · 1 unverified | report | |||||
| 157 | Shift-S | 47.8 | – | Paper | Code | 2022 | 6 of 7 ran · 1 unverified | report | |||||
| 158 | MogaNet-S (Semantic FPN) | 47.7 | 189 | – | Paper | Code | 2022 | 12 of 15 ran · 3 unverified | report | ||||
| 159 | SeMask (SeMask Swin-S FPN) | 47.63 | 56 | – | Paper | Code | 2021 | linked, not harvested | report | ||||
| 160 | ResNeSt-269 | 47.60 | – | Paper | Code | 2020 | 8 of 48 ran · 40 unverified | report | |||||
| 161 | UperNet Shuffle-T | 47.6 | – | Paper | Code | 2021 | 5 of 12 ran · 7 unverified | report | |||||
| 162 | CondNet(ResNest-101) | 47.54 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 163 | tiny-MOAT-3 (IN-1K pretraining, single scale) | 47.5 | 24 | – | Paper | Code | 2022 | linked, not harvested | report | ||||
| 164 | CondNet(ResNet-101) | 47.38 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 165 | DiNAT-Mini (UperNet) | 47.2 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 166 | DCNAS | 47.12 | – | Paper | – | 2020 | no code linked | report | |||||
| 167 | XCiT-S24/8 (Semantic-FPN) | 47.1 | – | Paper | Code | 2021 | 3 of 14 ran · 11 unverified | report | |||||
| 168 | ResNeSt-101 | 46.91 | – | Paper | Code | 2020 | 8 of 48 ran · 40 unverified | report | |||||
| 169 | XCiT-M24/8 (Semantic-FPN) | 46.9 | – | Paper | Code | 2021 | 3 of 14 ran · 11 unverified | report | |||||
| 170 | HamNet (ResNet-101) | 46.8 | – | Paper | Code | 2021 | 3 of 3 ran · 0 unverified | report | |||||
| 171 | Sequential Ensemble (DeepLabv3+) | 46.8 | – | Paper | – | 2022 | no code linked | report | |||||
| 172 | ConvNeXt-T | 46.7 | 60 | 939 | – | Paper | Code | 2022 | 54 of 80 ran · 26 unverified | report | |||
| 173 | VAN-Base (Semantic-FPN) | 46.7 | – | Paper | Code | 2022 | 0 of 6 ran · 6 unverified | report | |||||
| 174 | XCiT-S12/8 (UperNet) | 46.6 | – | Paper | Code | 2021 | 3 of 14 ran · 11 unverified | report | |||||
| 175 | GC ViT-T | 46.5 | 58 | 947 | – | Paper | Code | 2022 | 17 of 36 ran · 19 unverified | report | |||
| 176 | NAT-Mini | 46.4 | 50 | 900 | – | Paper | Code | 2022 | 1 of 1 ran · 0 unverified | report | |||
| 177 | Shift-T | 46.3 | – | Paper | Code | 2022 | 6 of 7 ran · 1 unverified | report | |||||
| 178 | DaViT-T | 46.3 | – | Paper | Code | 2022 | 8 of 15 ran · 7 unverified | report | |||||
| 179 | CPN(ResNet-101) | 46.27 | – | Paper | Code | 2020 | linked, not harvested | report | |||||
| 180 | MultiMAE (ViT-B) | 46.2 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 181 | DRAN(ResNet-101) | 46.18 | – | Paper | Code | 2020 | linked, not harvested | report | |||||
| 182 | PyConvSegNet-152 | 45.99 | 56.52 | – | Paper | Code | 2020 | linked, not harvested | report | ||||
| 183 | DNL | 45.97 | – | Paper | Code | 2020 | 1 of 3 ran · 2 unverified | report | |||||
| 184 | ACNet (ResNet-101) | 45.90 | – | Paper | – | 2019 | no code linked | report | |||||
| 185 | ACNet (ResNet-101) | 45.90 | – | Paper | – | 2019 | no code linked | report | |||||
| 186 | HRViT-b1 (SegFormer, SS) | 45.88 | 8.2 | 14.6 | – | Paper | Code | 2021 | linked, not harvested | report | |||
| 187 | OCR(HRNetV2-W48) | 45.66 | – | Paper | Code | 2019 | 4 of 9 ran · 5 unverified | report | |||||
| 188 | SPNet (ResNet-101) | 45.6 | – | Paper | Code | 2020 | 2 of 11 ran · 9 unverified | report | |||||
| 189 | Swin-T (UPerNet) MoBY | 45.58 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 190 | DAT-T (UperNet) | 45.54 | 60 | – | Paper | Code | 2022 | 8 of 10 ran · 2 unverified | report | ||||
| 191 | iBOT (ViT-S/16) | 45.4 | – | Paper | Code | 2021 | 0 of 1 ran · 1 unverified | report | |||||
| 192 | EANet (ResNet-101) | 45.33 | – | Paper | Code | 2021 | linked, not harvested | report | |||||
| 193 | OCR (ResNet-101) | 45.28 | – | Paper | Code | 2019 | 4 of 9 ran · 5 unverified | report | |||||
| 194 | Asymmetric ALNN | 45.24 | – | Paper | Code | 2019 | 0 of 3 ran · 3 unverified | report | |||||
| 195 | Light-Ham (VAN-Small, D=256) | 45.2 | 13.8 | 15.8 | – | Paper | Code | 2021 | 3 of 3 ran · 0 unverified | report | |||
| 196 | LaU-regression-loss | 45.02 | 56.32 | – | Paper | Code | 2019 | linked, not harvested | report | ||||
| 197 | PSPNet | 44.94 | 55.38 | – | Paper | Code | 2016 | 7 of 29 ran · 22 unverified | report | ||||
| 198 | tiny-MOAT-2 (IN-1K pretraining, single scale) | 44.9 | 13 | – | Paper | Code | 2022 | linked, not harvested | report | ||||
| 199 | CFNet(ResNet-101) | 44.89 | – | Paper | Code | 2019 | linked, not harvested | report | |||||
| 200 | EncNet | 44.65 | 55.67 | – | Paper | Code | 2018 | 1 of 8 ran · 7 unverified | report | ||||
| 201 | LaU-offset-loss | 44.55 | 56.41 | – | Paper | Code | 2019 | linked, not harvested | report | ||||
| 202 | EncNet + JPU | 44.34 | 55.84 | – | Paper | Code | 2019 | linked, not harvested | report | ||||
| 203 | SGR (ResNet-101) | 44.32 | – | Paper | Code | 2018 | linked, not harvested | report | |||||
| 204 | XCiT-S12/8 (Semantic-FPN) | 44.2 | – | Paper | Code | 2021 | 3 of 14 ran · 11 unverified | report | |||||
| 205 | Auto-DeepLab-L | 43.98 | – | Paper | Code | 2019 | 1 of 4 ran · 3 unverified | report | |||||
| 206 | PSANet (ResNet-101) | 43.77 | – | Paper | Code | 2018 | linked, not harvested | report | |||||
| 207 | DSSPN (ResNet-101) | 43.68 | – | Paper | – | 2018 | no code linked | report | |||||
| 208 | PSPNet (ResNet-152) | 43.51 | – | Paper | Code | 2016 | 7 of 29 ran · 22 unverified | report | |||||
| 209 | PSPNet (ResNet-101) | 43.29 | – | Paper | Code | 2016 | 7 of 29 ran · 22 unverified | report | |||||
| 210 | HRNetV2 | 43.2 | – | Paper | Code | 2019 | 3 of 18 ran · 15 unverified | report | |||||
| 211 | SeMask (SeMask Swin-T FPN) | 43.16 | 35 | – | Paper | Code | 2021 | linked, not harvested | report | ||||
| 212 | tiny-MOAT-1 (IN-1K pretraining, single scale) | 43.1 | 8 | – | Paper | Code | 2022 | linked, not harvested | report | ||||
| 213 | VAN-Small | 42.9 | 18 | – | Paper | Code | 2022 | 0 of 6 ran · 6 unverified | report | ||||
| 214 | PoolFormer-M48 | 42.7 | – | Paper | Code | 2021 | 2 of 3 ran · 1 unverified | report | |||||
| 215 | UperNet (ResNet-101) | 42.66 | – | Paper | Code | 2018 | 9 of 28 ran · 19 unverified | report | |||||
| 216 | tiny-MOAT-0 (IN-1K pretraining, single scale) | 41.2 | 6 | – | Paper | Code | 2022 | linked, not harvested | report | ||||
| 217 | RefineNet | 40.7 | – | Paper | Code | 2016 | 1 of 7 ran · 6 unverified | report | |||||
| 218 | FBNetV5 | 40.4 | – | Paper | – | 2021 | no code linked | report | |||||
| 219 | ConvMLP-L | 40 | – | Paper | Code | 2021 | 0 of 3 ran · 3 unverified | report | |||||
| 220 | ConvMLP-M | 38.6 | – | Paper | Code | 2021 | 0 of 3 ran · 3 unverified | report | |||||
| 221 | VAN-Tiny | 38.5 | 8 | – | Paper | Code | 2022 | 0 of 6 ran · 6 unverified | report | ||||
| 222 | A2MIM (ResNet-50) | 38.3 | – | Paper | Code | 2022 | linked, not harvested | report | |||||
| 223 | iBOT (ViT-B/16) (linear head) | 38.3 | – | Paper | Code | 2021 | 0 of 1 ran · 1 unverified | report | |||||
| 224 | SegFormer-B0 | 37.4 | 3.8 | ✓ | Paper | Code | 2021 | 48 of 86 ran · 38 unverified | report | ||||
| 225 | MUXNet-m + PPM | 35.8 | – | Paper | Code | 2020 | 3 of 5 ran · 2 unverified | report | |||||
| 226 | ConvMLP-S | 35.8 | – | Paper | Code | 2021 | 0 of 3 ran · 3 unverified | report | |||||
| 227 | MUXNet-m + C1 | 32.42 | – | Paper | Code | 2020 | 3 of 5 ran · 2 unverified | report | |||||
| 228 | DilatedNet | 32.31 | – | Paper | Code | 2015 | 2 of 8 ran · 6 unverified | report | |||||
| 229 | FCN | 29.39 | ✓ | Paper | Code | 2014 | 3 of 4 ran · 1 unverified | report | |||||
| 230 | SegNet | 21.64 | – | Paper | Code | 2015 | 9 of 44 ran · 35 unverified | report | |||||
| 231 | InternImage-H (M3I Pre-training) | 1310 | – | Paper | Code | 2022 | 2 of 4 ran · 2 unverified | report | |||||
| 232 | FastViT-MA36 | 44.6 | – | Paper | Code | 2023 | 1 of 5 ran · 4 unverified | report | |||||
| 233 | FastViT-SA36 | 42.9 | – | Paper | Code | 2023 | 1 of 5 ran · 4 unverified | report | |||||
| 234 | FastViT-SA24 | 41 | – | Paper | Code | 2023 | 1 of 5 ran · 4 unverified | report | |||||
| 235 | FastViT-SA12 | 38 | – | Paper | Code | 2023 | 1 of 5 ran · 4 unverified | report |
All 235 rows shown. 235 link to a paper page on this site; 18 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28
Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 156 rows have a graph line, from 67 distinct papers; 137 rows (59 papers) have at least one sample that ran. Counting each paper once: Syntology ran 572 of 1,238 samples; 666 unverified. Separately, 304 of those 1,238 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.
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