IslamicTurathBench: leaderboard
Classical Islamic scholarship benchmark: multiple-choice, passage-comprehension and closed-book open-ended questions across seven disciplines and three difficulty levels; ISTB score (%) is the mean of all 63 factorial cell means.
Metric: IslamicTurathBench Score (%). Source: github.com. Saturation forecast: Around February 2027. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 88.86 | #71 |
| 2 | GPT-5.1 | 82.61 | #145 |
| 3 | DeepSeek Reasoner | 77.63 | #254 |
| 4 | DeepSeek V3 Chat | 76.51 | #457 |
| 5 | Jais-2-70B-Chat | 72.96 | #430 |
| 6 | Fanar-C-2-27B | 63.37 | #637 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=islamicturathbench · How It Works · Data refreshed daily, snapshot 2026-11-05.