IntentGrasp - Gem Set - Smart Assistant: leaderboard
Metric: Instance F1 (%) on the Smart Assistant (SA) domain of the 470-item Gem Set, mean of four option-order runs, zero-shot, API default settings; higher is better. Source: arxiv.org. Saturation forecast: Around 2028. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 31.88 |
| 2 | GPT-5.4 | 19 |
| 3 | Gemini 3.1 Flash Lite (Preview) | 15.87 |
| 4 | Gemini 3.1 Pro (Preview) | 14.12 |
| 5 | Claude Sonnet 4.6 | 11.92 |
| 6 | Claude Opus 4.7 | 9.92 |
| 7 | GPT-5.4 Nano | 8.56 |
| 8 | GPT-5.4 Mini | 7.92 |
| 9 | Claude Haiku 4.5 | 3.2 |
Interactive version: theaggregate.ai/benchmark?slug=intentgrasp-gem-set-smart-assistant · How It Works · Data refreshed daily, snapshot 2026-10-07.