Benchmark Almanac

MOSE

DeAOT 59.4% tops its leaderboard
Not in model cards
Version 1

Models follow and segment objects in videos with crowds, occlusions, and objects leaving the screen. The benchmark compares masks against human annotations.

Made by
Henghui Ding and colleagues
Released
February 2023
Kind
Video understanding
Level
Everyday
Scored
Checked automatically
Answers
Open-ended
Tasks
All public
Measures
Capability
In and out
Video → Video, Images
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
17
Entries in its paper, 0 models we track

DeAOT tops the leaderboard with 59.4%

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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 DeAOT Mask initialization 59.4%
  2. 2 XMem Mask initialization 57.6%
  3. 3 AOT Mask initialization 57.2%
  4. 4 STCN Interactive, 60 seconds 56.8%
  5. 5 MiVOS Interactive, 60 seconds 51.4%
  6. 6 SWEM Mask initialization 50.9%
  7. 7 GIS Interactive, 60 seconds 50.2%
  8. 8 RDE Mask initialization 48.8%
  9. 9 STEm-Seg Zero-shot 46.9%
  10. 10 STM Interactive, 60 seconds 45.3%
  11. 11 ATNet Interactive, 60 seconds 44.5%
  12. 12 AGNN Zero-shot 43.7%
  13. 13 MANet Interactive, 60 seconds 43.6%
  14. 14 IPNet Interactive, 60 seconds 41.2%
  15. 15 RVOS Zero-shot 30.5%
  16. 16 FTMU Box initialization 23.8%
  17. 17 SiamMask Box initialization 22%

Source: The MOSE paper

#EntryScoreSettingDateSource
1 DeAOT 59.4% Mask initialization Official leaderboard
2 XMem 57.6% Mask initialization Official leaderboard
3 AOT 57.2% Mask initialization Official leaderboard
4 STCN 56.8% Interactive, 60 seconds Official leaderboard
5 MiVOS 51.4% Interactive, 60 seconds Official leaderboard
6 SWEM 50.9% Mask initialization Official leaderboard
7 GIS 50.2% Interactive, 60 seconds Official leaderboard
8 RDE 48.8% Mask initialization Official leaderboard
9 STEm-Seg 46.9% Zero-shot Official leaderboard
10 STM 45.3% Interactive, 60 seconds Official leaderboard
11 ATNet 44.5% Interactive, 60 seconds Official leaderboard
12 AGNN 43.7% Zero-shot Official leaderboard
13 MANet 43.6% Interactive, 60 seconds Official leaderboard
14 IPNet 41.2% Interactive, 60 seconds Official leaderboard
15 RVOS 30.5% Zero-shot Official leaderboard
16 FTMU 23.8% Box initialization Official leaderboard
17 SiamMask 22% Box initialization 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.

Versions

More in Audio and video

All 695 in Audio and video
  • Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy. Qwen 3.8 Max 88.7%
  • LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization. Gemini 4 Argon 91.7%
  • MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy. Qwen 3.8 Max 82.4%
  • MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy. Qwen 3.8 Max 90.8%
  • FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better. Phi-4-multimodal 4.00%
  • VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules. Qwen3.5-Omni-Plus 93.1%

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 benchmark’s paper (17 entries).
  • Pass Models scored 17 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom No scores yet.
Leaderboard import 3 pass
  • Pass Importer We read the board on henghuiding.github.io: 17 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