HighlightBench (Synthetic) - Aggregation and Comparison: leaderboard
Metric: Exact-match accuracy (%) on the aggregation and comparison task family of the synthetic subset (template-generated tables with controlled markups); higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash | 67.8 | #237 |
| 2 | Claude Sonnet 4 | 37.5 | #194 |
| 3 | MiniCPM-V-2.6 | 13.1 | #825 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=highlightbench-synthetic-aggregation-and-comparison · How It Works · Data refreshed daily, snapshot 2026-10-11.