ARFBench - Categorization: leaderboard
Metric: Accuracy (%) on the 104 Categorization questions (level shift, transient spike, seasonality, variance or trend change) of ARFBench, for the time series rendered as chart images (Matplotlib plots, three images for paired questions), with the series description and channel names as text; few-shot multiple-choice prompting with shuffled answer choices, temperature 0.05 (default for models without the setting), reasoning models at medium effort with 2,000 output tokens, blank or invalid answers counted wrong; higher is better. Source: arxiv.org. Saturation forecast: Around June 2028. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 63.5 |
| 2 | GPT-5 | 59.6 |
| 3 | GPT-4.1 | 56.7 |
| 4 | GPT-5.4 (Medium) | 56.7 |
| 5 | Gemini 3 Pro | 54.8 |
| 6 | Claude Sonnet 4.5 | 53.8 |
| 7 | GPT-4o | 51.9 |
Interactive version: theaggregate.ai/benchmark?slug=arfbench-categorization · How It Works · Data refreshed daily, snapshot 2026-10-07.