IntentGrasp - Gem Set - Writing: leaderboard

Metric: Instance F1 (%) on the Writing (W) 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 2029. 9 models tracked.

Top models

#ModelScore
1Gemini 3 Flash (Preview)20.17
2Gemini 3.1 Pro (Preview)12
3GPT-5.4 Mini5.83
4GPT-5.4 Nano5.17
5Gemini 3.1 Flash Lite (Preview)4.83
6GPT-5.44.5
7Claude Opus 4.73.83
8Claude Sonnet 4.63.5
9Claude Haiku 4.53.17

Interactive version: theaggregate.ai/benchmark?slug=intentgrasp-gem-set-writing · How It Works · Data refreshed daily, snapshot 2026-10-07.