Path-to-View on VIEWSUITE Path-to-View (All)
53.1AccuracyGPT-5.4 Pro
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-5.4 ProModel Category=Proprietary2026.05 | 53.1 | |
| Gemini 3.1 ProModel Category=Proprietary2026.05 | 48.8 | |
| GPT-5.4Model Category=Proprietary2026.05 | 47.8 | |
| Grok 4.20 BetaModel Category=Proprietary2026.05 | 46.1 | |
| GPT-5.1Model Category=Proprietary2026.05 | 43.9 | |
| Qwen3.5-397BModel Category=Open-Weight2026.05 | 39.7 | |
| Gemini 3 ProModel Category=Proprietary2026.05 | 37.8 | |
| Claude Opus 4.6Model Category=Proprietary2026.05 | 34.8 | |
| Qwen2.5-VL-7BModel Category=Open-Weight2026.05 | 29.5 | |
| Qwen2.5-VL-72BModel Category=Open-Weight2026.05 | 28.9 | |
| Kimi K2.5Model Category=Open-Weight2026.05 | 28.5 | |
| GLM-4.6VModel Category=Open-Weight2026.05 | 27.8 | |
| Qwen3-VL-32BModel Category=Open-Weight2026.05 | 27.4 | |
| Random ResponseModel Category=Baseline2026.05 | 23.3 |