Benchmark Almanac

EmbodiedCLUE-World

Seedance-2.0 77.2 tops its leaderboard Vidu Q3 Pro 76.2 leads the models we track
Not in model cards
Version Original

World models predict outcomes in embodied settings.

SuperCLUE reports scores in a public workbook.

Made by
SuperCLUE
Released
May 2026
Kind
Interactive world models
Level
Everyday
Scored
Checked automatically
Answers
Prediction
Measures
Capability
In and out
Video → Video
Interaction
Single turn
Run by
A host
Scores refreshed
November 20th, 2026
7
Entries on its own leaderboard, 5 models we track

Seedance-2.0 tops the leaderboard with 77.2 Vidu Q3 Pro leads with 76.2

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. The best model we track is Vidu Q3 Pro, at 76.2. This view shows only models we track, each at its best.

  1. 1 Seedance-2.0 API 77.2
  2. 2 Kling-3.0 API 77.1
  3. 3 Vidu Q3 Pro API 76.2
  4. 4 Runway Gen-4.5 API 76.1
  5. 5 Hailuo 2.3 API 74.0
  6. 6 Veo 3.1 Audio API 71.7
  7. 7 Wan 2.7 I2V API 68.4

Source: Official EmbodiedCLUE-World leaderboard

#EntryScoreSettingDateSource
1 Seedance-2.0 77.2 API Official leaderboard
2 Kling-3.0 77.1 API Official leaderboard
3 Vidu Q3 Pro 76.2 API Official leaderboard
4 Runway Gen-4.5 76.1 API Official leaderboard
5 Hailuo 2.3 74.0 API Official leaderboard
6 Veo 3.1 Audio 71.7 API Official leaderboard
7 Wan 2.7 I2V 68.4 API Official leaderboard
#ModelScoreSettingDateSource
1 Vidu Q3 Pro 76.2 API Official leaderboard
2 Runway Gen-4.5 76.1 API Official leaderboard
3 Hailuo 2.3 74.0 API Official leaderboard
4 Veo 3.1 Audio 71.7 API Official leaderboard
5 Wan 2.7 I2V 68.4 API Official leaderboard

Rankings by part

We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as 任务遵循, 物理交互 and 空间一致. Hailuo 2.3 leads 2 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

任务遵循

5 models we track have a score here.

  1. 1 Hailuo 2.3 ↑2 4.69
  2. 2 Vidu Q3 Pro ↓1 4.65
  3. 3 Wan 2.7 I2V ↑2 4.65
  4. 4 Runway Gen-4.5 ↓2 4.64
  5. 5 Veo 3.1 Audio ↓1 4.61

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

No dated values yet.

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 World models

All 53 in World models
  • AgentWorldBench In use Language world models predict the next environment observation after an agent action, in 7 domains from MCP and search to Android and desktop OS. A judge model scores each prediction against the real observation on five dimensions. GPT-5.4 58.3
  • IntPhys 2 In use Models watch simulated videos and judge whether each event is physically possible, testing object permanence, immutability, continuity and solidity. Scored by accuracy. Muse Spark Open 68.3%
  • QwenWorldBench Fading Models act as world models that simulate agent environments in 7 domains, such as terminal, web, and Android. A rubric judge scores five dimensions. Qwen3.7-Plus 62.1%
  • VPCT Fading Models view an image of a ball, ramps, and three buckets and predict which bucket the ball falls into. Scored by accuracy on 100 problems. Gemini 3 Pro 91%
  • ContPhy Fading Models infer physical properties such as mass and density of fluids and soft bodies from videos and predict their dynamics. Seed2.0 Pro 67.4%
  • RBench Only in system cards 650 image+text prompts for robot video generation (tasks and embodiments). MLLM judge scores task completion, physical plausibility, stability. Wan 2.6 60.7%

Data audit

7 of 8 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (7 entries).
  • Pass Models scored 5 models we track, plus 2 other entries on its leaderboard.
  • Note Newest score No scores yet.
  • Note Frontier models None of its models are on the capabilities index.
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on superclueai.com: 5 rows for models we track and 2 other entries at the last read.
  • 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