Benchmark Almanac

KITTI depth prediction

MD2E 7.28 tops its leaderboard
Not in model cards

Models estimate scene depth from road images. The evaluation compares model predictions with depth measurements collected from a moving vehicle.

Kind
Spatial reasoning and counting
Level
School
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
In and out
Images → Images
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
58
Entries on its own leaderboard, 0 models we track

MD2E tops the leaderboard with 7.28

Best score per entry from the benchmark's own leaderboard, read on November 21st, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

  1. 1 MD2E Eigen test split 7.28
  2. 2 ×Net Eigen test split 7.51
  3. 3 UniDepthV2 Eigen test split 7.74
  4. 4 OCEM Eigen test split 7.91
  5. 5 UniDepth Eigen test split 8.13
  6. 6 HyperDepth Eigen test split 9.16
  7. 7 MSFusion Eigen test split 9.37
  8. 8 DCDepth Eigen test split 9.60
  9. 9 NDDepth Eigen test split 9.62
  10. 10 IEBins Eigen test split 9.63
  11. 11 VA-DepthNet Eigen test split 9.84
  12. 12 DiffusionDepth-I Eigen test split 9.85
  13. 13 iDisc Eigen test split 9.89
  14. 14 MG Eigen test split 9.93
  15. 15 URCDC-Depth Eigen test split 10.0
  16. 16 BinsFormer Eigen test split 10.1
  17. 17 TrapNet Eigen test split 10.2
  18. 18 PixelFormer Eigen test split 10.3
  19. 19 RED-T Eigen test split 10.4
  20. 20 NeWCRFs Eigen test split 10.4
  21. 21 DepthFormer Eigen test split 10.7
  22. 22 ViP-DeepLab Eigen test split 10.8
  23. 23 SideRT Eigen test split 11.4
  24. 24 PWA Eigen test split 11.4
  25. 25 BANet Eigen test split 11.6
  26. 26 BTS Eigen test split 11.7
  27. 27 DL_61 (DORN) Eigen test split 11.8
  28. 28 RefinedMPL Eigen test split 11.8
  29. 29 DLE Eigen test split 11.8
  30. 30 PFANet Eigen test split 11.8
  31. 31 GAC Eigen test split 12.1
  32. 32 DL_SORD_SL Eigen test split 12.4
  33. 33 VNL Eigen test split 12.7
  34. 34 P3Depth Eigen test split 12.8
  35. 35 MS-DPT Eigen test split 12.8
  36. 36 DS-SIDENet_ROB Eigen test split 12.9
  37. 37 DL_SORD_SQ Eigen test split 13.0
  38. 38 PAP Eigen test split 13.1
  39. 39 CADepth-Net Eigen test split 13.3
  40. 40 VGG16-UNet Eigen test split 13.4
  41. 41 DORN_ROB Eigen test split 13.5
  42. 42 g2s Eigen test split 14.2
  43. 43 MT-SfMLearner Eigen test split 14.3
  44. 44 MLDA-Net Eigen test split 14.4
  45. 45 DABC_ROB Eigen test split 14.5
  46. 46 BTSREF_RVC Eigen test split 14.7
  47. 47 SDNet Eigen test split 14.7
  48. 48 APMoE_base_ROB Eigen test split 14.7
  49. 49 DiPE Eigen test split 14.8
  50. 50 CSWS_E_ROB Eigen test split 14.8
  51. 51 R-MSFM Eigen test split 15.1
  52. 52 HBC Eigen test split 15.2
  53. 53 SGDepth Eigen test split 15.3
  54. 54 DHGRL Eigen test split 15.5
  55. 55 GCNDepth Eigen test split 15.5
  56. 56 packnSFMHR_RVC Eigen test split 15.8
  57. 57 MultiDepth Eigen test split 16.1
  58. 58 LSIM Eigen test split 17.9

Source: Official KITTI depth prediction leaderboard

#EntryScoreSettingDateSource
1 MD2E 7.28 Eigen test split Official leaderboard
2 ×Net 7.51 Eigen test split Official leaderboard
3 UniDepthV2 7.74 Eigen test split Official leaderboard
4 OCEM 7.91 Eigen test split Official leaderboard
5 UniDepth 8.13 Eigen test split Official leaderboard
6 HyperDepth 9.16 Eigen test split Official leaderboard
7 MSFusion 9.37 Eigen test split Official leaderboard
8 DCDepth 9.60 Eigen test split Official leaderboard
9 NDDepth 9.62 Eigen test split Official leaderboard
10 IEBins 9.63 Eigen test split Official leaderboard
11 VA-DepthNet 9.84 Eigen test split Official leaderboard
12 DiffusionDepth-I 9.85 Eigen test split Official leaderboard
13 iDisc 9.89 Eigen test split Official leaderboard
14 MG 9.93 Eigen test split Official leaderboard
15 URCDC-Depth 10.0 Eigen test split Official leaderboard
16 BinsFormer 10.1 Eigen test split Official leaderboard
17 TrapNet 10.2 Eigen test split Official leaderboard
18 PixelFormer 10.3 Eigen test split Official leaderboard
19 RED-T 10.4 Eigen test split Official leaderboard
20 NeWCRFs 10.4 Eigen test split Official leaderboard
21 DepthFormer 10.7 Eigen test split Official leaderboard
22 ViP-DeepLab 10.8 Eigen test split Official leaderboard
23 SideRT 11.4 Eigen test split Official leaderboard
24 PWA 11.4 Eigen test split Official leaderboard
25 BANet 11.6 Eigen test split Official leaderboard
26 BTS 11.7 Eigen test split Official leaderboard
27 DL_61 (DORN) 11.8 Eigen test split Official leaderboard
28 RefinedMPL 11.8 Eigen test split Official leaderboard
29 DLE 11.8 Eigen test split Official leaderboard
30 PFANet 11.8 Eigen test split Official leaderboard
31 GAC 12.1 Eigen test split Official leaderboard
32 DL_SORD_SL 12.4 Eigen test split Official leaderboard
33 VNL 12.7 Eigen test split Official leaderboard
34 P3Depth 12.8 Eigen test split Official leaderboard
35 MS-DPT 12.8 Eigen test split Official leaderboard
36 DS-SIDENet_ROB 12.9 Eigen test split Official leaderboard
37 DL_SORD_SQ 13.0 Eigen test split Official leaderboard
38 PAP 13.1 Eigen test split Official leaderboard
39 CADepth-Net 13.3 Eigen test split Official leaderboard
40 VGG16-UNet 13.4 Eigen test split Official leaderboard
41 DORN_ROB 13.5 Eigen test split Official leaderboard
42 g2s 14.2 Eigen test split Official leaderboard
43 MT-SfMLearner 14.3 Eigen test split Official leaderboard
44 MLDA-Net 14.4 Eigen test split Official leaderboard
45 DABC_ROB 14.5 Eigen test split Official leaderboard
46 BTSREF_RVC 14.7 Eigen test split Official leaderboard
47 SDNet 14.7 Eigen test split Official leaderboard
48 APMoE_base_ROB 14.7 Eigen test split Official leaderboard
49 DiPE 14.8 Eigen test split Official leaderboard
50 CSWS_E_ROB 14.8 Eigen test split Official leaderboard
51 R-MSFM 15.1 Eigen test split Official leaderboard
52 HBC 15.2 Eigen test split Official leaderboard
53 SGDepth 15.3 Eigen test split Official leaderboard
54 DHGRL 15.5 Eigen test split Official leaderboard
55 GCNDepth 15.5 Eigen test split Official leaderboard
56 packnSFMHR_RVC 15.8 Eigen test split Official leaderboard
57 MultiDepth 16.1 Eigen test split Official leaderboard
58 LSIM 17.9 Eigen test split Official leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Vision

All 636 in Vision
  • MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. Qwen 3.6 Plus 86%
  • MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
  • CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
  • MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy. Seed2.1 Pro 90.7%
  • MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
  • MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy. Qwen 3.8 Max 85.9%

Data audit

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (58 entries).
  • Pass Models scored 58 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on cvlibs.net: 58 entries at the last read, none a model we track.
  • Pass Last read November 21st, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check