Eye-Q - Few-Shot CoT - Persian: leaderboard

Metric: Exact-Match Accuracy (%; three solved examples from the same subset, Persian subset). Source: arxiv.org. Saturation forecast: Around December 2026. 6 models tracked.

Top models

#ModelScore
1Gemini 2.5 Pro30.88
2Grok 4.1 Fast (Reasoning)19.04
3Gemini 2.5 Flash18.44
4GPT-5.2 (Non-reasoning)9.45
5Qwen 3 VL 235B A22B Instruct3.3
6Llama 4 Scout1.8

Interactive version: theaggregate.ai/benchmark?slug=eye-q-few-shot-cot-persian · How It Works · Data refreshed daily, snapshot 2026-09-25.