TRLawBench: leaderboard
Turkish legal benchmark testing legal knowledge, reasoning, and open-ended analysis over Turkish law tasks.
Metric: Stage 1 accuracy (imported). Source: www.betaspacestudio.com. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 32 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash | 85 | #245 |
| 2 | O4 Mini | 84.4 | #180 |
| 3 | Gemini 2.5 Pro | 83.8 | #152 |
| 4 | O3 | 81.3 | #124 |
| 5 | GPT-4.1 | 80 | #248 |
| 6 | Claude 3.7 Sonnet | 79.4 | #249 |
| 7 | GPT-4.1 Mini | 76.3 | #355 |
| 8 | Claude Sonnet 4 | 75.6 | #199 |
| 9 | Claude 3.7 Sonnet (Thinking) | 75.6 | #249 (Claude 3.7 Sonnet) |
| 10 | GPT-4o | 75 | #343 |
| 11 | Claude 3.5 Sonnet | 74.4 | #350 |
| 12 | DeepSeek Reasoner | 73.8 | #242 |
| 13 | Qwen 3 235B A22B | 72.5 | #313 |
| 14 | Mistral Large | 71.3 | #710 |
| 15 | Gemma 3 27B (IT) | 70 | #523 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=trlawbench · How It Works · Data refreshed daily, snapshot 2026-10-20.