TravelEval - Intercity Time Deviation: leaderboard
TravelEval (KDD 2026) asks an LLM for a complete multi-day travel plan and simulates it in a sandbox of 10 Chinese cities with real transport, hotel and attraction data. Scores are from Direct prompting over 1,150 queries. The board is the relative gap between planned and actual intercity travel time, a ratio where 0 is exact; lower is better.
Metric: Intercity Traffic Deviation (ITD): relative deviation of planned from actual intercity travel time (ratio, 0 is exact); Direct prompting (one pass, no tools), plans simulated in the TravelEval sandbox of 10 Chinese cities (real rail, flight, hotel and attraction data, queuing-time model, road-network distances) over 1,150 queries; lower is better. Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 0.14 | #250 |
| 2 | DeepSeek V3.1 (Non-reasoning) | 0.14 | #497 (DeepSeek V3.1) |
| 3 | GPT-5 Chat | 0.16 | #444 |
| 4 | Gemini 2.0 Flash | 0.17 | #704 |
| 5 | GPT-4o (2024-11-20) | 0.19 | #866 |
| 6 | GPT-4o Mini (2024-07-18) | 0.32 | #1510 |
| 7 | Qwen 3 8B | 0.33 | #1830 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=traveleval-intercity-time-deviation · How It Works · Data refreshed daily, snapshot 2026-11-13.