SuperCLUE-World
World models generate visual sequences.
SuperCLUE reports scores in a public workbook.
- Made by
- SuperCLUE
- Kind
- Video world models
- Level
- University
- Scored
- Graded by people
- Answers
- Open-ended
- Measures
- Capability
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
- In views
- Vision, Image, video and 3D generation
- Scores refreshed
- November 20th, 2026
Veo 3.1 Fast Audio leads with 74.6 Veo 3.1 Fast Audio leads with 74.6
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. This view shows only models we track, each at its best.
- 1
Veo 3.1 Fast Audio API 74.6
- 2 Sora 2 API 72.6
- 3
Wan 2.6 T2V API 71.3
- 4 Kling-2.6 API 69.4
- 5 Doubao-Seedance-1.5-pro API 65.0
- 6
Hailuo 2.3 API 63.9
- 7 PixVerse V5.5 API 63.5
- 8 Vidu Q2 API 62.6
Source: Official SuperCLUE-World leaderboard
- 1
Veo 3.1 Fast Audio API 74.6
- 2 Sora 2 API 72.6
- 3
Wan 2.6 T2V API 71.3
- 4
Hailuo 2.3 API 63.9
Source: Official SuperCLUE-World leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 74.6 | API | Official leaderboard | |
| 2 | Sora 2 | 72.6 | API | Official leaderboard | |
| 3 | | 71.3 | API | Official leaderboard | |
| 4 | Kling-2.6 | 69.4 | API | Official leaderboard | |
| 5 | Doubao-Seedance-1.5-pro | 65.0 | API | Official leaderboard | |
| 6 | | 63.9 | API | Official leaderboard | |
| 7 | PixVerse V5.5 | 63.5 | API | Official leaderboard | |
| 8 | Vidu Q2 | 62.6 | API | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 74.6 | API | Official leaderboard | |
| 2 | Sora 2 | 72.6 | API | Official leaderboard | |
| 3 | | 71.3 | API | Official leaderboard | |
| 4 | | 63.9 | API | Official leaderboard |
Rankings by part
We read 8 parts from the benchmark’s own leaderboard in 2 groups: 基础能力 and 应用能力. Each ranks models on one part of the test. Veo 3.1 Fast Audio leads 5 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
基础能力
4 models we track have a score here.
- 1
Veo 3.1 Fast Audio 76.9
- 2 Sora 2 71.8
- 3
Wan 2.6 T2V 70.3
- 4
Hailuo 2.3 64.7
应用能力
4 models we track have a score here.
- 1 Sora 2 ↑1 73.4
- 2
Veo 3.1 Fast Audio ↓1 72.3
- 3
Wan 2.6 T2V 72.2
- 4
Hailuo 2.3 63.0
几何 一致性
4 models we track have a score here.
- 1
Veo 3.1 Fast Audio 79.0
- 2
Hailuo 2.3 ↑2 78.6
- 3
Wan 2.6 T2V 75.7
- 4 Sora 2 ↓2 75.1
视觉 真实感
4 models we track have a score here.
- 1
Wan 2.6 T2V ↑2 83.9
- 2
Veo 3.1 Fast Audio ↓1 78.0
- 3 Sora 2 ↓1 76.7
- 4
Hailuo 2.3 67.5
相机控制
4 models we track have a score here.
- 1
Veo 3.1 Fast Audio 76.3
- 2 Sora 2 65.6
- 3
Wan 2.6 T2V 65.1
- 4
Hailuo 2.3 64.3
动态 性
4 models we track have a score here.
- 1
Veo 3.1 Fast Audio 74.2
- 2 Sora 2 69.9
- 3
Wan 2.6 T2V 56.7
- 4
Hailuo 2.3 48.5
商业领域
4 models we track have a score here.
- 1 Sora 2 ↑1 75.7
- 2
Wan 2.6 T2V ↑1 68.5
- 3
Veo 3.1 Fast Audio ↓2 68.1
- 4
Hailuo 2.3 68.0
创意设计
4 models we track have a score here.
- 1
Veo 3.1 Fast Audio 76.6
- 2
Wan 2.6 T2V ↑1 76.0
- 3 Sora 2 ↓1 71.1
- 4
Hailuo 2.3 58.0
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.
More 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 (8 entries).
- Pass Models scored 4 models we track, plus 4 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: 4 rows for models we track and 4 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.