EuroEval Italian NLU - ScaLA IT: leaderboard

Metric: Linguistic acceptability Score (%). Source: euroeval.com. 435 models tracked.

Top models

#ModelScore
1Gemini 3 Pro (Preview)80.08
2Gemini 3.1 Flash Lite (Preview)79.29
3Gemma 4 31B (IT)75.49
4Claude Sonnet 4.674.36
5GPT-5.266.85
6Gemma 4 26B A4B (IT)66.16
7Llama 3.1 70B64.33
8GPT-5.4 Mini (High)62.31
9Qwen 3.5 9B62.14
10Grok 4.1 Fast (Reasoning)61.32
11GPT-5.4 Mini (Medium)60.16
12gemma-3-27B-pt57.15
13Qwen 3.5 4B57.06
14GLM-4.7 Flash55.95
15Ministral-3-3B-Instruct-251253.69

Interactive version: theaggregate.ai/benchmark?slug=euroeval-italian-nlu-scala-it · How It Works · Data refreshed daily, snapshot 2026-09-05.