EKI Estonian LLM - Idioms: leaderboard
EKI idioms benchmark: 300 fill-in-the-blank tasks restoring Estonian-specific, universal and ambiguous figurative expressions in exactly the right grammatical form, with accepted variants vetted by EKI linguists. Accuracy 0-100, higher is better.
Metric: Accuracy (%). Source: moodupuu.eki.ee. Saturation forecast: Around February 2027. 100 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 74 | #6 |
| 2 | GPT-5.5 | 73.67 | #33 |
| 3 | Gemini 3.8 Flash | 73.33 | #30 |
| 4 | Gemini 3.1 Pro (Preview) | 72.33 | #64 |
| 5 | GPT-5.6 Sol | 70.67 | #22 |
| 6 | GPT-6 Sol | 70.67 | #19 |
| 7 | Gemini 3.7 Flash | 70.33 | #24 |
| 8 | Gemini 4 Argon | 70 | #9 |
| 9 | GPT-6.1 Terra | 69.33 | #15 |
| 10 | Claude Fable 5 | 65 | #20 |
| 11 | GPT-6 Luna | 64.67 | #61 |
| 12 | GPT-5.6 Terra | 64.33 | #40 |
| 13 | Gemini 3.5 Flash | 64 | #66 |
| 14 | GPT-5.6 Luna | 64 | #79 |
| 15 | Gemini 3.6 Flash | 63.33 | #57 |
No result here: #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=eki-estonian-llm-idioms · How It Works · Data refreshed daily, snapshot 2026-11-05.