EuroEval Bulgarian: leaderboard
Bulgarian-language EuroEval all-task suite spanning NLU, knowledge, common-sense reasoning, reading comprehension, summarization, and related tasks.
Metric: Average Score (%). Source: euroeval.com. Saturation forecast: Around November 2026. 219 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 74.47 | #66 |
| 2 | Gemini 3 Flash (Preview) (Thinking) | 73.3 | #82 (Gemini 3 Flash (Preview)) |
| 3 | Gemini 3 Flash (Preview) (Non-reasoning) | 73.15 | #82 (Gemini 3 Flash (Preview)) |
| 4 | GPT-5 | 72.19 | #98 |
| 5 | Claude Sonnet 4.5 (Thinking) | 72 | #142 (Claude Sonnet 4.5) |
| 6 | Qwen 3.5 27B FP8 | 71.66 | #164 |
| 7 | GPT-5.4 Mini (2026-03-17) (High) | 71.12 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 8 | Gemini 2.5 Flash (Thinking) | 71.04 | #245 (Gemini 2.5 Flash) |
| 9 | Gemini 2.5 Pro | 70.78 | #152 |
| 10 | Gemini 3.1 Flash Lite (Preview) | 70.38 | #194 |
| 11 | Claude Sonnet 4.6 | 70.3 | #90 |
| 12 | GPT-5 Mini (2025-08-07) | 69.35 | #172 |
| 13 | Gemma 4 31B (IT) | 68.9 | #176 |
| 14 | GPT-4.1 | 67.66 | #248 |
| 15 | Qwen 3.5 35B A3B Base | 67.35 | #310 |
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 · How It Works · Data refreshed daily, snapshot 2026-10-20.