TemporalBench - MIMIC T2: leaderboard
TemporalBench MIMIC clinical slice for T2 event-informed forecasting over patient measurements.
Metric: Accuracy (%). Source: huggingface.co. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | TimeClaw (deepseek-v4-pro) | 33.33 | |
| 2 | TimeCopilot (deepseek-v4-pro) | 30 | |
| 3 | CAMEL (deepseek-chat) | 24.82 | |
| 4 | TimeCopilot (deepseek-chat) | 23.66 | |
| 5 | TimeSeries Scientist (deepseek-chat) | 23.4 | |
| 6 | TimeSeries Scientist (gpt-4o) | 23.4 | |
| 7 | AgentScope (deepseek-chat) | 22.7 | |
| 8 | MetaGPT (deepseek-chat) | 22.7 | |
| 9 | Single LLM (deepseek-chat) | 22.67 | |
| 10 | AgentScope (gpt-4o) | 21.28 |
Interactive version: theaggregate.ai/benchmark?slug=temporalbench-mimic-t2 · How It Works · Data refreshed daily, snapshot 2026-10-20.