Perception on MMStar
83.6ScoreGemini-3-Flash
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-3-Flash2026.06 | 83.6 | |
| GPT-5.12026.06 | 71.6 | |
| Qwen3-VL + LOCUSSize=4B2026.06 | 69.7 | |
| MiMo-VL + LOCUSSize=7B2026.06 | 68.9 | |
| Claude-3.7-SonnetModel Category=Closed-Source2026.03 | 68.8 | |
| Qwen3-VLSize=4B2026.06 | 66.5 | |
| MiMo-VLSize=7B2026.06 | 66.3 | |
| MM-Eureka-7BModel Category=Multimodal Reasoning Models2026.03 | 65.2 | |
| GPT-4oModel Category=Closed-Source2026.03 | 65.1 | |
| ThinkLite-VLModel Category=Multimodal Reasoning Models2026.03 | 65 | |
| Vision-R1Model Category=Multimodal Reasoning Models2026.03 | 64.8 | |
| VLAA-Thinker-7BModel Category=Multimodal Reasoning Models2026.03 | 64.2 | |
| Vision-SR1Model Category=Multimodal Reasoning Models2026.03 | 64.1 | |
| AVAR-ThinkerModel Category=Our model2026.03 | 64.1 | |
| InternVL2.5-8BModel Category=Open-Source General Models2026.03 | 63.2 | |
| Qwen2.5-VL + LOCUSSize=7B2026.06 | 63.1 | |
| Qwen2.5-VL-7BModel Category=Open-Source General Models2026.03 | 62.1 | |
| LLaVA-OneVision-7BModel Category=Open-Source General Models2026.03 | 61.7 | |
| Qwen2.5-VLSize=7B2026.06 | 61.5 | |
| Mulberry-7BModel Category=Multimodal Reasoning Models2026.03 | 61.3 | |
| OpenVLThinkerModel Category=Multimodal Reasoning Models2026.03 | 59.5 | |
| R1-OneVisionModel Category=Multimodal Reasoning Models2026.03 | 52.2 | |
| Llama-3.2-11B-Vision-InstructModel Category=Open-Source General Models2026.03 | 49.8 |