TemporalBench - T2: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | TimeClaw (deepseek-v4-pro) | 38.22 |
| 2 | AgentScope (deepseek-chat) | 31.37 |
| 3 | Single LLM (deepseek-chat) | 30.19 |
| 4 | CAMEL (deepseek-chat) | 29.85 |
| 5 | Single LLM (gpt-4o) | 29.84 |
| 6 | MetaGPT (gpt-4o) | 27.33 |
| 7 | AgentScope (gpt-4o) | 26.53 |
| 8 | TimeSeries Scientist (deepseek-chat) | 26.53 |
| 9 | TimeSeries Scientist (gpt-4o) | 26.53 |
| 10 | CAMEL (gpt-4o) | 26.18 |
Interactive version: theaggregate.ai/benchmark?slug=temporalbench-t2 · How It Works · Data refreshed daily, snapshot 2026-09-05.