EuroEval Bulgarian NLU: leaderboard
Bulgarian-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: NLU Average Score (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 219 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 Mini (2026-03-17) (High) | 73.19 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 2 | Gemini 3 Pro (Preview) | 72.77 | #66 |
| 3 | Gemini 3 Flash (Preview) (Thinking) | 72.62 | #82 (Gemini 3 Flash (Preview)) |
| 4 | Gemini 3.1 Flash Lite (Preview) | 72.42 | #194 |
| 5 | Gemini 2.5 Flash (Thinking) | 72.35 | #245 (Gemini 2.5 Flash) |
| 6 | Gemini 3 Flash (Preview) (Non-reasoning) | 71.62 | #82 (Gemini 3 Flash (Preview)) |
| 7 | Gemma 4 31B (IT) | 71.39 | #176 |
| 8 | GPT-5.2 | 70.81 | #110 |
| 9 | Gemini 2.5 Flash Lite (Thinking) | 70.77 | #422 (Gemini 2.5 Flash Lite) |
| 10 | Claude Sonnet 4.5 (Thinking) | 70.64 | #142 (Claude Sonnet 4.5) |
| 11 | Claude Sonnet 4.6 | 70.6 | #90 |
| 12 | Qwen 3.5 35B A3B Base | 70.6 | #310 |
| 13 | GPT-4.1 | 70.01 | #248 |
| 14 | GPT-5 | 69.93 | #98 |
| 15 | Ministral-3-14B-Reasoning-2512 | 69.66 | #540 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-bulgarian-nlu · How It Works · Data refreshed daily, snapshot 2026-10-20.