IntentGrasp - Gem Set - Customer Support: leaderboard

Metric: Instance F1 (%) on the Customer Support (CS) 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)24.73
2Claude Opus 4.78.92
3Gemini 3.1 Flash Lite (Preview)7.45
4GPT-5.47.15
5GPT-5.4 Nano5.12
6Gemini 3.1 Pro (Preview)4.83
7Claude Sonnet 4.64.5
8Claude Haiku 4.52.5
9GPT-5.4 Mini1

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