Multimodal Perception on V*
89.6AccuracyTACO (Ours-7B)
Evaluation results
| Method | Links | |
|---|---|---|
| TACO (Ours-7B)Model Category=Code-tool / visual-agent models2026.06 | 89.6 | |
| PyVision-RL-7BModel Category=Code-tool / visual-agent models2026.06 | 88.7 | |
| Mini-o3-7BModel Category=Code-tool / visual-agent models2026.06 | 88.2 | |
| Qwen3-VL-8BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.06 | 86.4 | |
| DeepEyes-7BModel Category=Code-tool / visual-agent models2026.06 | 85.6 | |
| CodeV-RL-7BModel Category=Code-tool / visual-agent models2026.06 | 84.8 | |
| Pixel-Reasoner-7BModel Category=Code-tool / visual-agent models2026.06 | 84.3 | |
| Thyme-7BModel Category=Code-tool / visual-agent models2026.06 | 82.2 | |
| DeepEyes-v2-7BModel Category=Code-tool / visual-agent models2026.06 | 81.8 | |
| Qwen2.5-VL-32BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.06 | 81.2 | |
| Gemini-2.5-ProModel Category=Closed-source models2026.06 | 79.1 | |
| MathCoder-VL-8BModel Category=Code-tool / visual-agent models2026.06 | 77.5 | |
| Qwen2.5-VL-7BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.06 | 76.4 | |
| LLaVA-OneVision-7BModel Category=Open-source MLLMs (no visual tool)2026.06 | 75.4 | |
| GPT-5Model Category=Closed-source models2026.06 | 72.5 | |
| InternVL3-8BModel Category=Open-source MLLMs (no visual tool)2026.06 | 70.2 | |
| GPT-4oModel Category=Closed-source models2026.06 | 67.5 |