PRISM-Bench (audio and video generation)
Models generate speech, music, or environmental sounds with video. A judge compares outputs with reference clips across 35 criteria, including synchronization and prompt following.
The authors publish aggregate scores and six example clips. Redistribution rights limit access to the full audiovisual benchmark.
- Made by
- Shanghai Artificial Intelligence Laboratory and Meituan
- Released
- May 2026
- Kind
- Video generation
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- Public, plus a private test set
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
- Scores refreshed
- November 20th, 2026
Dreamina Seedance 2.0 leads with 91.9 Dreamina Seedance 2.0 leads with 91.9
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. The board reports no overall score, so this is the plain mean of its three Final Scores. Off-Screen uses nine dimensions and the other sets use twelve. 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
Dreamina Seedance 2.0 91.9
- 2 Kling v3 Omni 84.5
- 3
Veo 3.1 Audio 79.7
- 4
Veo 3 Audio 77.7
- 5 Sora 2 73.6
- 6 Ovi 46.3
- 7 LTX-2 45.9
- 8 MoVA 39.4
Source: Official PRISM-Bench (audio and video generation) leaderboard
- 1
Dreamina Seedance 2.0 91.9
- 2
Veo 3.1 Audio 79.7
- 3
Veo 3 Audio 77.7
- 4 Sora 2 73.6
Source: Official PRISM-Bench (audio and video generation) leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 91.9 | Official leaderboard | ||
| 2 | Kling v3 Omni | 84.5 | Official leaderboard | ||
| 3 | | 79.7 | Official leaderboard | ||
| 4 | | 77.7 | Official leaderboard | ||
| 5 | Sora 2 | 73.6 | Official leaderboard | ||
| 6 | Ovi | 46.3 | Official leaderboard | ||
| 7 | LTX-2 | 45.9 | Official leaderboard | ||
| 8 | MoVA | 39.4 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 91.9 | Official leaderboard | ||
| 2 | | 79.7 | Official leaderboard | ||
| 3 | | 77.7 | Official leaderboard | ||
| 4 | Sora 2 | 73.6 | Official leaderboard |
Rankings by part
We read 3 parts from the benchmark’s own leaderboard: Mixed-AB, On-Screen and Off-Screen. Each ranks models on one part of the test. Dreamina Seedance 2.0 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
Mixed-AB
4 models we track have a score here.
- 1
Dreamina Seedance 2.0 99.9
- 2 Sora 2 ↑2 89.0
- 3
Veo 3.1 Audio ↓1 88.0
- 4
Veo 3 Audio ↓1 83.6
On-Screen
4 models we track have a score here.
- 1
Dreamina Seedance 2.0 97.7
- 2
Veo 3.1 Audio 84.1
- 3
Veo 3 Audio 83.5
- 4 Sora 2 69.8
Off-Screen
4 models we track have a score here.
- 1
Dreamina Seedance 2.0 78.1
- 2
Veo 3.1 Audio 67.0
- 3
Veo 3 Audio 66.1
- 4 Sora 2 61.8
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 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 Public tasks, plus a private test set.
- 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 huggingface.co: 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.