WorldTravel-Bench (Text): leaderboard
Humanlaya real-world travel-planning feasibility benchmark, text setting: share of generated itineraries that are actually feasible; human performance reference 77.9%.
Metric: Feasibility Rate (%). Source: humanlaya.com. Saturation forecast: Around March 2028. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 | 32.67 | #115 |
| 2 | Claude Opus 4.5 | 21.33 | #87 |
| 3 | GPT-5.1 | 16.67 | #145 |
| 4 | Gemini 3 Pro | 14.67 | #82 |
| 5 | DeepSeek V3.2 | 14 | #210 |
| 6 | Claude Sonnet 4.5 | 10 | #148 |
| 7 | Gemini 2.5 Pro | 8 | #159 |
| 8 | GLM-4.6 | 8 | #265 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=worldtravel-bench-text · How It Works · Data refreshed daily, snapshot 2026-11-05.