IntentGrasp - Gem Set - Policy Making: leaderboard

The Policy Making domain of the IntentGrasp (2026) Gem Set, hard intent-understanding items that every open-weight model failed in its first run. The score is instance F1 in percent, averaged over four option orders; higher is better.

Metric: Instance F1 (%) on the Policy Making (PM) 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 February 2027. 9 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Flash (Preview)39.77#156
2Gemini 3.1 Flash Lite (Preview)26.44#419
3Claude Opus 4.720.42#86
4Gemini 3.1 Pro (Preview)20.08#102
5Claude Sonnet 4.617.23#179
6GPT-5.412.31#138
7GPT-5.4 Nano2.35#653
8GPT-5.4 Mini1.14#409
9Claude Haiku 4.50#587

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=intentgrasp-gem-set-policy-making · How It Works · Data refreshed daily, snapshot 2026-11-22.