- MathVista 82.5
- MathVerse 81
- MathVision 61.9
- MMMU 72.2
- VLMs Are Blind 77.3
- VisuLogic 31.1
- VisualPuzzle 52.5
- ChartQA 87.1
- AI2D (transparent) 95.3
- DocVQA (val) 93.6
- OCRBench 858 (out of 1000) 85.8
- CharXiv-RQ (DQ 90.3) 56.9
- SimpleVQA 53.1
- HallusionBench 60.4
- MMStar 71.1
- BLINK 62.8
- CV-Bench 83.8
- CountBench 89
- TempCompass 75.1
ERNIE-4.5-VL-28B-A3B-Thinking
ERNIE-4.5-VL-28B-A3B-Thinking places in the top three on AI2D.
19
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Baidu Nov 2025 ERNIE-4.5-VL-28B-A3B-Thinking
Open weights Small
No Epoch Capabilities Index score yet
- SimpleVQA 53.1% #21 /23
- HallusionBench 60.4% #18 /23
- BLINK 62.8% #16 /28
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 | |
|---|---|---|---|---|
| HallusionBench | 60.4% | #18 of 23 | Model card | |
| SimpleVQA | 53.1% | #21 of 23 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AI2D | 95.3% | #3 of 44 | Model card | |
| MathVerse | 81% | #4 of 12 | Model card | |
| MathVista | 82.5% | #22 of 64 | Model card | |
| ChartQA | 87.1% | #17 of 37 | Model card | |
| MMMU | 72.2% | #51 of 97 | Model card | |
| DocVQA | 93.6% | #18 of 33 | Model card | |
| BLINK | 62.8% | #16 of 28 | Model card | |
| OCRBench | 85.8% | #18 of 28 | Model card | |
| MMStar | 71.1% | #21 of 30 | Model card | |
| CountBench | 89% | #14 of 20 | Model card | |
| CVBench | 83.8% | #8 of 11 | Model card | |
| MATH-Vision (MathVision) | 61.9% | #32 of 42 | Model card | |
| TempCompass | 75.1% | #6 of 7 | Model card | |
| VisuLogic | 31.1% | #6 of 7 | Model card | |
| CharXiv | 56.9% | #59 of 67 | Model card | |
| VLMs Are Blind | 77.3% | #8 of 8 | Model card | |
| VisualPuzzles | 52.5% | #3 of 3 | Model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.