EKI Estonian LLM - Trivia Hard: leaderboard
EKI open-answer trivia benchmark (Mälumäng): 800 harder questions about Estonia, 80 in each of ten topics, graded by exact match with a GPT-5.4-nano judge for non-matching answers. Accuracy 0-100, higher is better; retired models were never run on it.
Metric: Accuracy (%). Source: moodupuu.eki.ee. Saturation forecast: Around July 2027. 92 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 50.75 | #64 |
| 2 | Gemini 4 Argon | 50.62 | #9 |
| 3 | Gemini 3.8 Flash | 50.25 | #30 |
| 4 | Gemini 3.7 Flash | 47 | #24 |
| 5 | GPT-6 Astra | 45.75 | #6 |
| 6 | GPT-5.6 Sol | 43.88 | #22 |
| 7 | Gemini 3.5 Flash | 42.25 | #66 |
| 8 | Gemini 3.6 Flash | 41.25 | #57 |
| 9 | GPT-5.5 | 38.38 | #33 |
| 10 | GPT-6 Sol | 37.12 | #19 |
| 11 | GPT-6.1 Terra | 30.38 | #15 |
| 12 | Claude Fable 5 | 28.5 | #20 |
| 13 | Claude Fable 5.1 | 27.12 | #10 |
| 14 | Gemini 3 Flash (Preview) | 26.75 | #90 |
| 15 | Gemini 2.5 Pro | 26 | #159 |
No result here: #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=eki-estonian-llm-trivia-hard · How It Works · Data refreshed daily, snapshot 2026-11-05.