IntentGrasp - General: leaderboard

The General domain of the IntentGrasp (2026) All Set, intent-understanding items drawn from public intent datasets and posed as multiple choice with one or more correct intents. The score is instance F1 in percent, averaged over option-order shuffles; higher is better.

Metric: Instance F1 (%) on the General (G) domain of the 12,909-item All Set, mean over option-order shuffles, zero-shot at temperature 0; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemma 4 31B (IT) [4bit]69.87#1824
2Olmo-3.1-32B-Instruct [4bit]66.46
3gemma-4-E4B-it62.5#1577
4Qwen3-32B [4bit]59.99#1296
5Llama 3.3 70B Instruct [4bit]59.65#1275
6Qwen 3 8B59.08#1890
7Gemma 4 E2B (IT)53.95#2532
8Qwen 3 4B53.56#2379
9Llama 3.1 8B Instruct50.31#4655
10OLMo 3 7B Instruct47.69#6953
11Llama 3.2 3B Instruct43.01#7311

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

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