RT-QA - One-Hop (L2): leaderboard
Metric: Accuracy (%) on 1-hop questions that first resolve the target entity, mean of the six days; Chinese real-time questions whose answers come from executing a verified scraping workflow at the moment of inference, asked daily from January 10 to 15, 2026; a DeepResearch-style agent with Serper search and Jina page reading, the current timestamp in the prompt, temperature 0.6; a GPT-5.2 judge checks semantic equivalence; higher is better. Source: arxiv.org. Saturation forecast: Around September 2028. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Seed 1.8 | 45.6 |
| 2 | GLM-4.7 | 41.9 |
| 3 | GPT-5.2 | 38 |
| 4 | DeepSeek V3.2 | 34 |
| 5 | Gemini 3 Pro | 33.2 |
Interactive version: theaggregate.ai/benchmark?slug=rt-qa-one-hop-l2 · How It Works · Data refreshed daily, snapshot 2026-10-07.