GeoFlip
Models generate videos from paired scenes where a geometry change reverses the expected physical outcome. Model judges check the event and the scene structure.
- Made by
- Yuqi Li and colleagues
- Released
- October 2026
- Kind
- Video generation
- Level
- School
- Scored
- Graded by a model
- Answers
- Open-ended
- Measures
- Capability
- Subject
- Physical sciences
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
- In views
- Vision, Image, video and 3D generation
- Last model added
- July 2025
- Scores refreshed
- November 20th, 2026
Wan2.2-I2V-A14B leads with 3.17 Wan2.2-I2V-A14B leads with 3.17
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Table 2 ranks models by the mean of their ranks across three dimensions. A lower mean rank is better. 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
Wan2.2-I2V-A14B 3.17
- 2
MiniMax H3 3.67
- 3 LingBot-World-Fast 4.33
- 4 LTX-Video-2B 5.83
- 5 HunyuanVideo-1.5-I2V 6.00
- 6 Wan2.1-I2V-14B-480P 6.00
- 7 Cosmos3-Super 6.17
- 8 VBVR-Wan2.2 7.00
- 9 DynamiCrafter-1024 7.67
- 10 SVD-XT 9.67
- 11 CogVideoX-5B-I2V 10.2
- 12 SkyReels-V2-I2V-14B 10.2
- 13 I2VGen-XL 12.8
- 14 CogVideoX1.5-5B-I2V 13.3
- 15 HunyuanVideo-I2V-720P 14.3
- 16 SkyReels-V1-I2V 15.7
Source: The GeoFlip paper
- 1
Wan2.2-I2V-A14B 3.17
- 2
MiniMax H3 3.67
Source: The GeoFlip paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 3.17 | Jul 2025 | Official leaderboard | |
| 2 | | 3.67 | Official leaderboard | ||
| 3 | LingBot-World-Fast | 4.33 | Official leaderboard | ||
| 4 | LTX-Video-2B | 5.83 | Official leaderboard | ||
| 5 | HunyuanVideo-1.5-I2V | 6.00 | Official leaderboard | ||
| 6 | Wan2.1-I2V-14B-480P | 6.00 | Official leaderboard | ||
| 7 | Cosmos3-Super | 6.17 | Official leaderboard | ||
| 8 | VBVR-Wan2.2 | 7.00 | Official leaderboard | ||
| 9 | DynamiCrafter-1024 | 7.67 | Official leaderboard | ||
| 10 | SVD-XT | 9.67 | Official leaderboard | ||
| 11 | CogVideoX-5B-I2V | 10.2 | Official leaderboard | ||
| 12 | SkyReels-V2-I2V-14B | 10.2 | Official leaderboard | ||
| 13 | I2VGen-XL | 12.8 | Official leaderboard | ||
| 14 | CogVideoX1.5-5B-I2V | 13.3 | Official leaderboard | ||
| 15 | HunyuanVideo-I2V-720P | 14.3 | Official leaderboard | ||
| 16 | SkyReels-V1-I2V | 15.7 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 3.17 | Jul 2025 | Official leaderboard | |
| 2 | | 3.67 | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official GeoFlip 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
- 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 benchmark’s paper (16 entries).
- Pass Models scored 2 models we track, plus 14 other entries on its leaderboard.
- Warning Newest score The newest came 16 months ago, in July 2025, for Wan2.2-I2V-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: 2 rows for models we track and 14 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.