Eye-Q - Partial Character Reveal - Persian: leaderboard
Metric: Exact-Match Accuracy (%; 25% of the answer's characters revealed, Persian subset). Source: arxiv.org. Saturation forecast: Around December 2026. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 43.03 |
| 2 | Grok 4.1 Fast (Reasoning) | 31.93 |
| 3 | Gemini 2.5 Flash | 18.29 |
| 4 | GPT-5.2 (Non-reasoning) | 13.79 |
| 5 | Llama 4 Scout | 4.8 |
| 6 | Qwen 3 VL 235B A22B Instruct | 3.15 |
Interactive version: theaggregate.ai/benchmark?slug=eye-q-partial-character-reveal-persian · How It Works · Data refreshed daily, snapshot 2026-09-25.