Kosmos-2
Kosmos-2 places in the top three on SEED-Bench-2.
13
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Microsoft Jun 2023 Kosmos-2
Open weights Small
No Epoch Capabilities Index score yet
- SEED-Bench 50% #7 /9
- MM-Vet 23.7% #160 /164
- SEED-Bench-2-Plus 25.3% #135 /136
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 | 0.90% | #178 of 178 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SEED-Bench-2 | 52.4% | #2 of 5 | Official leaderboard | |
| SEED-Bench | 50% | #7 of 9 | Official leaderboard | |
| MM-Vet | 23.7% | #160 of 164 | Official leaderboard | |
| CoreCognition | 34.9% | #117 of 119 | Official leaderboard | |
| CCBench (MMBench Chinese Culture) | 1.40% | #150 of 152 | Official leaderboard | |
| SEED-Bench-2-Plus | 25.3% | #135 of 136 | Official leaderboard | |
| MMStar | 24.9% | #172 of 173 | Official leaderboard | |
| MMBench-EN v1.1 | 1.20% | #178 of 178 | Official leaderboard | |
| MMBench | 1.10% | #155 of 155 | Official leaderboard | |
| MMMU | 24.4% | #136 of 136 | Official leaderboard | |
| A-Bench | 33.2% | #116 of 116 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 21.0 | #172 of 173 | Official leaderboard |