Benchmark Almanac

DIODE

DPT 0.18 tops its leaderboard
Not in model cards

Models estimate depth in indoor and outdoor scenes. The dataset provides depth measurements from a laser scanner for comparison.

Released
August 2019
Kind
Spatial reasoning and counting
Level
School
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Images → Images
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
18
Entries on its own leaderboard, 0 models we track

DPT tops the leaderboard with 0.18

Best score per entry from the benchmark's own leaderboard, read on November 20th, 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 DPT Training Data=1.4M 0.18
  2. 2 GenPercept Training Data=90K, diffusion-based=true 0.23
  3. 3 FE2E Training Data=71K, Pre-trained Stable Diffusion=false 0.23
  4. 4 Lotus-D Training Data=59K, diffusion-based=true 0.23
  5. 5 Lotus-G Training Data=59K, diffusion-based=true 0.23
  6. 6 Iris Training Data=59K + 100K, diffusion-based=true 0.24
  7. 7 HDN Training Data=300K 0.25
  8. 8 DepthAnything Training Data=62.6M 0.26
  9. 9 DepthAnything V2 Training Data=62.6M 0.27
  10. 10 LeRes Training Data=354K 0.27
  11. 11 Diffusion-E2E-FT Training Data=74K, diffusion-based=true 0.30
  12. 12 Marigold v1.1 Training Data=74K, Pre-trained Stable Diffusion=true 0.30
  13. 13 Marigold(LCM) Training Data=74K, diffusion-based=true 0.31
  14. 14 Marigold Training Data=74K, diffusion-based=true 0.31
  15. 15 MiDaS Training Data=2M 0.33
  16. 16 GeoWizard Training Data=280K, diffusion-based=true 0.34
  17. 17 Omnidata Training Data=12.2M 0.34
  18. 18 DiverseDepth Training Data=320K 0.38

Source: Official DIODE leaderboard

#EntryScoreSettingDateSource
1 DPT 0.18 Training Data=1.4M Official leaderboard
2 GenPercept 0.23 Training Data=90K, diffusion-based=true Official leaderboard
3 FE2E 0.23 Training Data=71K, Pre-trained Stable Diffusion=false Official leaderboard
4 Lotus-D 0.23 Training Data=59K, diffusion-based=true Official leaderboard
5 Lotus-G 0.23 Training Data=59K, diffusion-based=true Official leaderboard
6 Iris 0.24 Training Data=59K + 100K, diffusion-based=true Official leaderboard
7 HDN 0.25 Training Data=300K Official leaderboard
8 DepthAnything 0.26 Training Data=62.6M Official leaderboard
9 DepthAnything V2 0.27 Training Data=62.6M Official leaderboard
10 LeRes 0.27 Training Data=354K Official leaderboard
11 Diffusion-E2E-FT 0.30 Training Data=74K, diffusion-based=true Official leaderboard
12 Marigold v1.1 0.30 Training Data=74K, Pre-trained Stable Diffusion=true Official leaderboard
13 Marigold(LCM) 0.31 Training Data=74K, diffusion-based=true Official leaderboard
14 Marigold 0.31 Training Data=74K, diffusion-based=true Official leaderboard
15 MiDaS 0.33 Training Data=2M Official leaderboard
16 GeoWizard 0.34 Training Data=280K, diffusion-based=true Official leaderboard
17 Omnidata 0.34 Training Data=12.2M Official leaderboard
18 DiverseDepth 0.38 Training Data=320K 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 (18 entries).
  • Pass Models scored 18 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 sota2.com: 18 entries at the last read, none a model we track.
  • Pass Last read November 20th, 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