Benchmark Almanac

PhyRubric-Bench

Seedance 2.5 44.2 tops its leaderboard Veo 3.1 Audio 36.2 leads the models we track
Not in model cards

Models generate videos of physical events. A judge checks caption-specific requirements against evidence at particular times in each clip.

The authors compare ten video generators across five physical domains.

Released
November 2026
Kind
Video generation
Level
School
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Video
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Last model added
July 2025
Scores refreshed
November 20th, 2026
10
Entries in its paper, 3 models we track

Seedance 2.5 tops the leaderboard with 44.2 Veo 3.1 Audio leads with 36.2 among models released by July 2025

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. The best model we track is Veo 3.1 Audio, at 36.2. This view shows only models we track, each at its best.

The newest model with a score here came out in July 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 Seedance 2.5 44.2
  2. 2 Veo 3.1 Audio 36.2
  3. 3 HunyuanVideo-1.5-8B 14.9
  4. 4 Wan2.2-T2V-A14B 14.9
  5. 5 Cosmos-Predict2.5-14B 12.2
  6. 6 LTX-2.3-22B 11.5
  7. 7 Wan2.2-TI2V-5B 7.87
  8. 8 Cosmos-Predict2.5-2B 6.69
  9. 9 CogVideoX1.5-5B 5.27
  10. 10 LTX-Video-2B 4.07

Source: The PhyRubric-Bench paper

#EntryScoreSettingDateSource
1 Seedance 2.5 44.2 Official leaderboard
2 Veo 3.1 Audio 36.2 Official leaderboard
3 HunyuanVideo-1.5-8B 14.9 Official leaderboard
4 Wan2.2-T2V-A14B 14.9 Jul 2025 Official leaderboard
5 Cosmos-Predict2.5-14B 12.2 Official leaderboard
6 LTX-2.3-22B 11.5 Official leaderboard
7 Wan2.2-TI2V-5B 7.87 Jul 2025 Official leaderboard
8 Cosmos-Predict2.5-2B 6.69 Official leaderboard
9 CogVideoX1.5-5B 5.27 Official leaderboard
10 LTX-Video-2B 4.07 Official leaderboard
#ModelScoreSettingDateSource
1 Veo 3.1 Audio 36.2 Official leaderboard
2 Wan2.2-T2V-A14B 14.9 Jul 2025 Official leaderboard
3 Wan2.2-TI2V-5B 7.87 Jul 2025 Official leaderboard

Rankings by part

We read 5 parts from the benchmark’s paper, each ranking models on one part of the test, such as Mechanics, Optics and Heat. Veo 3.1 Audio leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

Mechanics

3 models we track have a score here.

  1. 1 Veo 3.1 Audio 44.0
  2. 2 Wan2.2-T2V-A14B 21.3
  3. 3 Wan2.2-TI2V-5B 9.33

Scores over time

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

In the benchmark’s paper
6.00 8.00 10.0 12.0 14.0 16.0 6.00 8.00 10.0 12.0 14.0 16.0

Source: Official PhyRubric-Bench 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

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 16 months ago, in July 2025, for Wan2.2-T2V-A14B. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the benchmark’s paper (10 entries).
  • Pass Models scored 3 models we track, plus 7 other entries on its leaderboard.
  • Warning Newest score The newest came 16 months ago, in July 2025, for Wan2.2-T2V-A14B.
  • 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 arxiv.org: 3 rows for models we track and 7 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