TemporalBench - PSML T1: leaderboard
TemporalBench PSML process-systems slice for T1 contextual understanding of industrial time series.
Metric: Accuracy (%). Source: huggingface.co. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Single LLM (deepseek-chat) | 74 | |
| 2 | MetaGPT (deepseek-chat) | 71.5 | |
| 3 | TimeCopilot (deepseek-chat) | 71 | |
| 4 | AgentScope (deepseek-chat) | 68.5 | |
| 5 | CAMEL (gpt-4o) | 68.5 | |
| 6 | MetaGPT (gpt-4o) | 67.5 | |
| 7 | Single LLM (gpt-4o) | 67.5 | |
| 8 | AgentScope (gpt-4o) | 66 | |
| 9 | CAMEL (deepseek-chat) | 65 | |
| 10 | TimeClaw (deepseek-v4-pro) | 56 |
Interactive version: theaggregate.ai/benchmark?slug=temporalbench-psml-t1 · How It Works · Data refreshed daily, snapshot 2026-10-20.