Qwen2.5-VL-3B
Qwen2.5-VL-3B.
18
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Qwen Jan 2025 Qwen2.5-VL-3B
Open weights Small
No Epoch Capabilities Index score yet
- VisualPuzzles 31.2% #19 /29
- A-Bench 74.9% #35 /116
- Math-VR 5.30% #24 /26
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 | 76.5% | #71 of 178 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ScreenSpot-Pro | 16.1% | #8 of 11 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CCBench (MMBench Chinese Culture) | 62% | #37 of 152 | Official leaderboard | |
| KRETA | 71.8% | #5 of 17 | Official leaderboard | |
| A-Bench | 74.9% | #35 of 116 | Official leaderboard | |
| MMBench | 79.1% | #62 of 155 | Official leaderboard | |
| MMStar | 56.3% | #73 of 173 | Official leaderboard | |
| MMBench-EN v1.1 | 77% | #80 of 178 | Official leaderboard | |
| CharXiv | 31.3% | #42 of 77 | Official leaderboard | |
| MMMU | 53.1% | #77 of 136 | Official leaderboard | |
| VisualPuzzles | 31.2% | #19 of 29 | Official leaderboard | |
| VHELM | 40.7 | #29 of 41 | Official leaderboard | |
| MMMU-Pro | 31.6% | #41 of 57 | Official leaderboard | |
| MATH-Vision (MathVision) | 21.2% | #87 of 105 | Official leaderboard | |
| ERQA | 32.6% | #15 of 17 | Official leaderboard | |
| Math-VR | 5.30% | #24 of 26 | Official leaderboard | |
| RoboRewardBench | 1.61 | #19 of 19 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OpenVLM Leaderboard | 64.5 | #61 of 173 | Official leaderboard |