Multimodal Reasoning on MM-Vista
81.4AccuracyGPT-5
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5Model Category=Closed-source models2026.06 | 81.4 | |
| Gemini-2.5-ProModel Category=Closed-source models2026.06 | 80.9 | |
| Qwen3-VL-8BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.06 | 77.2 | |
| TACO (Ours-7B)Model Category=Code-tool / visual-agent models2026.06 | 76.3 | |
| MathCoder-VL-8BModel Category=Code-tool / visual-agent models2026.06 | 73.6 | |
| Qwen2.5-VL-32BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.06 | 72.2 | |
| DeepEyes-v2-7BModel Category=Code-tool / visual-agent models2026.06 | 71.9 | |
| CodeV-RL-7BModel Category=Code-tool / visual-agent models2026.06 | 71.8 | |
| PyVision-RL-7BModel Category=Code-tool / visual-agent models2026.06 | 71.4 | |
| Pixel-Reasoner-7BModel Category=Code-tool / visual-agent models2026.06 | 71.2 | |
| InternVL3-8BModel Category=Open-source MLLMs (no visual tool)2026.06 | 70.4 | |
| DeepEyes-7BModel Category=Code-tool / visual-agent models2026.06 | 70.1 | |
| Thyme-7BModel Category=Code-tool / visual-agent models2026.06 | 70 | |
| Qwen2.5-VL-7BModel Category=Open-source MLLMs (no visual tool), Instruct variant=true2026.06 | 68.2 | |
| Mini-o3-7BModel Category=Code-tool / visual-agent models2026.06 | 67.2 | |
| Unsilencing Latent ReasoningBackbone=VLAA Thinking-7B2026.05 | 65.7 | |
| GPT-4oModel Category=Closed-source models2026.06 | 63.4 | |
| DMLRBackbone=VLAA Thinking-7B2026.05 | 62.9 | |
| ICoTBackbone=VLAA Thinking-7B2026.05 | 61.4 | |
| VanillaBackbone=VLAA Thinking-7B2026.05 | 61.1 | |
| CCoTBackbone=VLAA Thinking-7B2026.05 | 60.5 | |
| MCoTBackbone=VLAA Thinking-7B2026.05 | 59.6 | |
| LLaVA-OneVision-7BModel Category=Open-source MLLMs (no visual tool)2026.06 | 58.6 | |
| DMLRBackbone=R1 OneVision-7B2026.05 | 58 | |
| Unsilencing Latent ReasoningBackbone=R1 OneVision-7B2026.05 | 57.3 | |
| ICoTBackbone=R1 OneVision-7B2026.05 | 55.6 | |
| CCoTBackbone=R1 OneVision-7B2026.05 | 53.4 | |
| MCoTBackbone=R1 OneVision-7B2026.05 | 52.5 | |
| VanillaBackbone=R1 OneVision-7B2026.05 | 51.2 |