ORAV Bench
Models combine reference images, sounds, and videos into a requested audiovisual scene. A judge compares pairs of outputs against each reference's assigned role.
- Made by
- Tsinghua University and Tencent Hunyuan
- Released
- September 2026
- Kind
- Video generation
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision, Audio and speech, Image, video and 3D generation
- Scores refreshed
- November 20th, 2026
Dreamina Seedance 2.5 leads with 0.70 Dreamina Seedance 2.5 leads with 0.70
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. This view shows only models we track, each at its best.
- 1
Dreamina Seedance 2.5 0.70
- 2
Dreamina Seedance 2.0 0.48
- 3
MiniMax H3 0.15
- 4
Wan 3.0 -0.56
- 5 Kling-v3-Omni -0.77
Source: The ORAV Bench paper
- 1
Dreamina Seedance 2.5 0.70
- 2
Dreamina Seedance 2.0 0.48
- 3
MiniMax H3 0.15
- 4
Wan 3.0 -0.56
Source: The ORAV Bench paper
| # | Entry | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 0.70 0.6 to 0.9 | Official leaderboard | ||
| 2 | | 0.48 0.4 to 0.6 | Official leaderboard | ||
| 3 | | 0.15 ±0.1 | Official leaderboard | ||
| 4 | | -0.56 ±0.1 | Official leaderboard | ||
| 5 | Kling-v3-Omni | -0.77 -0.9 to -0.6 | Official leaderboard |
| # | Model | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 0.70 0.6 to 0.9 | Official leaderboard | ||
| 2 | | 0.48 0.4 to 0.6 | Official leaderboard | ||
| 3 | | 0.15 ±0.1 | Official leaderboard | ||
| 4 | | -0.56 ±0.1 | Official leaderboard |
Rankings by part
We read 9 parts from the benchmark’s paper in 3 groups: image, video and audio. Each ranks models on one part of the test. Dreamina Seedance 2.5 leads 7 of them, and MiniMax H3 leads the other 2. Arrows show how far a model moved from its overall rank among the models we track.
Subject
4 models we track have a score here.
- 1
Dreamina Seedance 2.5 68.9%
- 2
Dreamina Seedance 2.0 62.8%
- 3
MiniMax H3 53.4%
- 4
Wan 3.0 35.5%
Prop
4 models we track have a score here.
- 1
Dreamina Seedance 2.5 64.1%
- 2
Dreamina Seedance 2.0 61%
- 3
MiniMax H3 56.4%
- 4
Wan 3.0 38.9%
Scene
4 models we track have a score here.
- 1
Dreamina Seedance 2.5 68.6%
- 2
Dreamina Seedance 2.0 61.9%
- 3
MiniMax H3 56.9%
- 4
Wan 3.0 34.7%
Style
4 models we track have a score here.
- 1
Dreamina Seedance 2.5 65.5%
- 2
Dreamina Seedance 2.0 63.5%
- 3
MiniMax H3 57.5%
- 4
Wan 3.0 36.5%
Motion
4 models we track have a score here.
- 1
Dreamina Seedance 2.5 71%
- 2
Dreamina Seedance 2.0 65.7%
- 3
MiniMax H3 42.4%
- 4
Wan 3.0 30.1%
Camera movement
4 models we track have a score here.
- 1
MiniMax H3 ↑2 65.9%
- 2
Dreamina Seedance 2.5 ↓1 65.4%
- 3
Dreamina Seedance 2.0 ↓1 64.9%
- 4
Wan 3.0 25%
VFX
4 models we track have a score here.
- 1
Dreamina Seedance 2.5 82.6%
- 2
Dreamina Seedance 2.0 71.4%
- 3
MiniMax H3 58.3%
- 4
Wan 3.0 30%
Speech
4 models we track have a score here.
- 1
MiniMax H3 ↑2 73.6%
- 2
Dreamina Seedance 2.5 ↓1 68%
- 3
Dreamina Seedance 2.0 ↓1 60%
- 4
Wan 3.0 40.9%
Music
4 models we track have a score here.
- 1
Dreamina Seedance 2.5 76.6%
- 2
Dreamina Seedance 2.0 51%
- 3
Wan 3.0 ↑1 40.4%
- 4
MiniMax H3 ↓1 39.9%
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 All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the benchmark’s paper (5 entries).
- Pass Models scored 4 models we track, plus 1 other entry 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 arxiv.org: 4 rows for models we track and 1 other entry 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.