Math & Reasoning on MathVista mini
82.9Overall ScoreQwen3.5-Omni
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3.5-OmniOpen-Source=false, Size=Flash2026.04 | 82.9 | |
| Nemotron 3 Nano OmniOpen-Source=true, Size=30B-A3B, Reasoning=on2026.04 | 82.8 | |
| Qwen3.5-4BModel=Qwen3.5-4B2026.05 | 82.3 | |
| Bee-8BOpenness=fully open, Model Size=8B, Training Strategy=RL2025.10 | 81.4 | |
| Keye-VLOpenness=semi-open, Model Size=8B2025.10 | 80.7 | |
| Qwen3-OmniOpen-Source=true, Size=30B-A3B2026.04 | 80 | |
| Bee-8BOpenness=fully open, Model Size=8B, Training Strategy=SFT2025.10 | 78.6 | |
| InternVL3.5Openness=semi-open, Model Size=8B2025.10 | 78.4 | |
| Nemotron Nano V2 VLOpen-Source=true, Size=12B, Reasoning=on2026.04 | 75.5 | |
| InternVL3.5-20B-A4BModel=InternVL3.5-20B-A4B2026.05 | 73.5 | |
| InternVL3.5-4BModel=InternVL3.5-4B2026.05 | 72.8 | |
| Nemotron 3 Nano OmniOpen-Source=true, Size=30B-A3B, Reasoning=off2026.04 | 71.9 | |
| Nemotron Nano V2 VLOpen-Source=true, Size=12B, Reasoning=off2026.04 | 69 | |
| Qwen2.5-VLOpenness=semi-open, Model Size=7B2025.10 | 68.2 | |
| ZAYA1-VL-8B-A1BModel=ZAYA1-VL-8B-A1B2026.05 | 64 | |
| Qwen3-VL-4BModel=Qwen3-VL-4B2026.05 | 63.6 | |
| LLaVA OneVisionOpenness=fully open, Model Size=7B2025.10 | 63.2 | |
| Qwen2.5-VL-3BModel=Qwen2.5-VL-3B2026.05 | 63.2 | |
| PLM-3BModel=PLM-3B2026.05 | 61.5 | |
| InternVL3.5-2BModel=InternVL3.5-2B2026.05 | 61.4 | |
| DeepSeek-VL2-16B-A2.4BModel=DeepSeek-VL2-16B-A2.4B2026.05 | 61.2 | |
| Molmo2-4BModel=Molmo2-4B2026.05 | 56.5 | |
| Qwen3.5-2BModel=Qwen3.5-2B2026.05 | 52.9 | |
| Qwen3-VL-2BModel=Qwen3-VL-2B2026.05 | 51.8 | |
| MolmoOpenness=fully open, Model Size=7B, Training Strategy=Distilled (-D)2025.10 | 51.6 | |
| MolmoE-8B-A1BModel=MolmoE-8B-A1B2026.05 | 39.1 |