Benchmark Almanac

SuperCLUE-Video

Doubao-Seedance-2.0 69.2 tops its leaderboard Wan 2.6 T2V 54.3 leads the models we track
Not in model cards

Models generate videos from Chinese prompts. A judging model examines video frames and scores them against criteria for the prompt.

Prompts include scenes specific to Chinese culture.

Made by
SuperCLUE team
Released
February 2024
Kind
Video generation
Level
School
Scored
Graded by a model
Answers
Open-ended
Measures
Capability
Language
Chinese
In and out
Text → Video
Interaction
Single turn
Refresh
One fixed set
Run by
A host
Last model added
February 2025
Scores refreshed
November 20th, 2026
32
Entries on its own leaderboard, 6 models we track

Doubao-Seedance-2.0 tops the leaderboard with 69.2 Wan 2.6 T2V leads with 54.3 among models released by February 2025

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 Wan 2.6 T2V, at 54.3. This view shows only models we track, each at its best.

The newest model with a score here came out in February 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Doubao-Seedance-2.0 网页获取 69.2
  2. 2 Veo-3.1-Generate-Preview API 65.7
  3. 3 HappyHorse-1.0-T2V API 63.8
  4. 4 PixVerse V6 API 62.3
  5. 5 Kling-3.0 网页获取 61.5
  6. 6 Vidu Q3 API 60.6
  7. 7 Doubao-Seedance-1.5-pro API 56.6
  8. 8 Kling-2.6 API 55.9
  9. 9 Veo-3.0-Generate-Preview API 55.4
  10. 10 Vidu Q2 API 55.0
  11. 11 Wan 2.6 T2V API 54.3
  12. 12 Hailuo-02 API 51.7
  13. 13 Kling-Video-O1 API 50.4
  14. 14 Doubao-Seedance-1.0-pro API 49.1
  15. 15 Sora 2 Pro API 48.0
  16. 16 Ray 3 API 47.7
  17. 17 Hailuo 2.3 API 46.8
  18. 18 Kling-2.1-Master API 44.8
  19. 19 PixVerse V5.5 API 42.7
  20. 20 Pika 2.2 网页获取 40.0
  21. 21 Sora 网页获取 38.0
  22. 22 Vidu Q1 API 36.7
  23. 23 Wan2.1-T2V-14B 本地部署 33.0
  24. 24 PixVerse V4.5 API 32.7
  25. 25 Wanx2.1-t2v-plus API 32.5
  26. 26 Longcat-Video API 32.1
  27. 27 Mochi 本地部署 29.0
  28. 28 CogVideoX-5B 本地部署 25.7
  29. 29 Open-Sora-v2 本地部署 25.5
  30. 30 Step-Video-T2V 本地部署 25.3
  31. 31 Cosmos-1.0-Diffusion-14B-Text2World 本地部署 24.4
  32. 32 Hunyuan Video 本地部署 22.3

Source: Official SuperCLUE-Video leaderboard

#EntryScoreSettingDateSource
1 Doubao-Seedance-2.0 69.2 网页获取 Official leaderboard
2 Veo-3.1-Generate-Preview 65.7 API Official leaderboard
3 HappyHorse-1.0-T2V 63.8 API Official leaderboard
4 PixVerse V6 62.3 API Official leaderboard
5 Kling-3.0 61.5 网页获取 Official leaderboard
6 Vidu Q3 60.6 API Official leaderboard
7 Doubao-Seedance-1.5-pro 56.6 API Official leaderboard
8 Kling-2.6 55.9 API Official leaderboard
9 Veo-3.0-Generate-Preview 55.4 API Official leaderboard
10 Vidu Q2 55.0 API Official leaderboard
11 Wan 2.6 T2V 54.3 API Official leaderboard
12 Hailuo-02 51.7 API Official leaderboard
13 Kling-Video-O1 50.4 API Official leaderboard
14 Doubao-Seedance-1.0-pro 49.1 API Official leaderboard
15 Sora 2 Pro 48.0 API Official leaderboard
16 Ray 3 47.7 API Official leaderboard
17 Hailuo 2.3 46.8 API Official leaderboard
18 Kling-2.1-Master 44.8 API Official leaderboard
19 PixVerse V5.5 42.7 API Official leaderboard
20 Pika 2.2 40.0 网页获取 Official leaderboard
21 Sora 38.0 网页获取 Official leaderboard
22 Vidu Q1 36.7 API Official leaderboard
23 Wan2.1-T2V-14B 33.0 本地部署 Feb 2025 Official leaderboard
24 PixVerse V4.5 32.7 API Official leaderboard
25 Wanx2.1-t2v-plus 32.5 API Official leaderboard
26 Longcat-Video 32.1 API Official leaderboard
27 Mochi 29.0 本地部署 Official leaderboard
28 CogVideoX-5B 25.7 本地部署 Official leaderboard
29 Open-Sora-v2 25.5 本地部署 Official leaderboard
30 Step-Video-T2V 25.3 本地部署 Official leaderboard
31 Cosmos-1.0-Diffusion-14B-Text2World 24.4 本地部署 Official leaderboard
32 Hunyuan Video 22.3 本地部署 Official leaderboard
#ModelScoreSettingDateSource
1 Wan 2.6 T2V 54.3 API Official leaderboard
2 Sora 2 Pro 48.0 API Official leaderboard
3 Ray 3 47.7 API Official leaderboard
4 Hailuo 2.3 46.8 API Official leaderboard
5 Sora 38.0 网页获取 Official leaderboard
6 Wan2.1-T2V-14B 33.0 本地部署 Feb 2025 Official leaderboard

Rankings by part

We read 9 parts from the benchmark’s own leaderboard in 2 groups: 通用能力榜单 and 场景应用榜单. Each ranks models on one part of the test. Wan 2.6 T2V 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.

通用能力

6 models we track have a score here.

  1. 1 Wan 2.6 T2V 53.2
  2. 2 Ray 3 ↑1 48.3
  3. 3 Hailuo 2.3 ↑1 47.6
  4. 4 Sora 2 Pro ↓2 43.0
  5. 5 Sora 38.3
  6. 6 Wan2.1-T2V-14B 37.2

Scores over time

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

On the benchmark’s own leaderboard
33.0 2025 33.0 2025

Source: Official SuperCLUE-Video 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 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

7 of 9 checks pass, 2 warnings. 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 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (32 entries).
  • Pass Models scored 6 models we track, plus 26 other entries on its leaderboard.
  • Warning Newest score The newest came 21 months ago, in February 2025, for Wan2.1-T2V-14B.
  • 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: 6 rows for models we track and 26 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