TimeSage-EV: leaderboard
Overall evaluation score combining accuracy, coverage, faithfulness and quality on TimeSage-EV. Reads only the Overall tier in Table 12, never the easier per-tier maximum.
Metric: Overall (%). Source: arxiv.org. Saturation forecast: Around January 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 85.99 | #79 |
| 2 | Qwen 3.5 397B A17B | 84.24 | #146 |
| 3 | Claude Sonnet 4.6 | 82.38 | #90 |
| 4 | Kimi K2.6 | 77.9 | #104 |
| 5 | Gemma 4 31B | 75.31 | #200 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=timesage-ev · How It Works · Data refreshed daily, snapshot 2026-10-20.