TemporalBench - PSML T2: leaderboard
TemporalBench PSML process-systems slice for T2 event-informed forecasting questions.
Metric: Accuracy (%). Source: huggingface.co. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | TimeSeries Scientist (deepseek-chat) | 26.67 | |
| 2 | TimeSeries Scientist (gpt-4o) | 26.67 | |
| 3 | AgentScope (gpt-4o) | 24.67 | |
| 4 | Single LLM (deepseek-chat) | 24.67 | |
| 5 | TimeClaw (deepseek-v4-pro) | 23.33 | |
| 6 | TimeCopilot (deepseek-chat) | 22.22 | |
| 7 | MetaGPT (gpt-4o) | 21.09 | |
| 8 | Single LLM (gpt-4o) | 20.67 | |
| 9 | AgentScope (deepseek-chat) | 20.51 | |
| 10 | MetaGPT (deepseek-chat) | 19.73 |
Interactive version: theaggregate.ai/benchmark?slug=temporalbench-psml-t2 · How It Works · Data refreshed daily, snapshot 2026-10-20.