Short-answer Visual Question Answering on AI2D
81.3ScoreIntern-VL-2.5-4B
Evaluation Results
| Method | Links | |
|---|---|---|
| Intern-VL-2.5-4BModel Category=Autoregressive2026.04 | 81.3 | |
| Qwen2.5-VL-3BModel Category=Autoregressive2026.04 | 80.8 | |
| Fast-dVLMModel Category=Diffusion, Decoding Strategy=MDM2026.04 | 79.7 | |
| Fast-dVLMModel Category=Diffusion, Decoding Strategy=spec.2026.04 | 79.7 | |
| LLaDA-VModel Category=Diffusion2026.04 | 77.8 | |
| DimpleModel Category=Diffusion2026.04 | 74.4 | |
| LaViDaModel Category=Diffusion2026.04 | 70 | |
| MiniCPM-V-2 (3B)Model Category=Autoregressive2026.04 | 65 | |
| VILA-1.5-3BModel Category=Autoregressive2026.04 | 58 |