Ovis2-16B
Ovis2-16B ranks first on Ego3D-Bench.
Alibaba International Digital Commerce (AIDC-AI) Released February 10th, 2025 Family Ovis2 The lab's model card Compare with others 17
Benchmarks with a score
1
Ranked first, out of five or more models
3
In the top three
Alibaba International Digital Commerce (AIDC-AI) Feb 2025 Ovis2-16B
Open weights Mid-size
No Epoch Capabilities Index score yet
- MM-Vet 68.4% #23 /164
- SEED-Bench-2-Plus 72.1% #3 /136
- A-Bench 79.4% #5 /116
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMBench-CN v1.1 | 85.5% | #14 of 178 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SEED-Bench-2-Plus | 72.1% | #3 of 136 | Official leaderboard | |
| A-Bench | 79.4% | #5 of 116 | Official leaderboard | |
| MMBench | 87.2% | #7 of 155 | Official leaderboard | |
| MMBench-EN v1.1 | 85.8% | #13 of 178 | Official leaderboard | |
| Ego3D-Bench | 7.40 | #1 of 11 | Official leaderboard | |
| MM-Vet | 68.4% | #23 of 164 | Official leaderboard | |
| MMStar | 67.2% | #25 of 173 | Official leaderboard | |
| EmbodiedBench EB-Navigation | 47.7% | #3 of 17 | Official leaderboard | |
| All-Angles Bench | 53.2% | #4 of 18 | Official leaderboard | |
| CCBench (MMBench Chinese Culture) | 62.4% | #36 of 152 | Official leaderboard | |
| MMMU | 59.6% | #57 of 136 | Official leaderboard | |
| EmbodiedBench EB-Habitat | 32% | #11 of 17 | Official leaderboard | |
| EmbodiedBench EB-ALFRED | 16.3% | #12 of 17 | Official leaderboard | |
| MATH-Vision (MathVision) | 30.1% | #75 of 105 | Official leaderboard | |
| EmbodiedBench EB-Manipulation | 11.3% | #13 of 17 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 73.3 | #20 of 173 | Official leaderboard |