Polish EQ-Bench — leaderboard
Polish adaptation of EQ-Bench: emotional intelligence benchmark testing LLMs on understanding and predicting emotional dynamics in Polish-language scenarios.
Metric: EQ-Bench Score. Source: huggingface.co. Status: saturated. 102 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Mistral Large 2 (Jul) | 78.07 |
| 2 | GPT-4 Turbo | 77.77 |
| 3 | Mistral Large 2 (Nov) Instruct (2411) | 77.29 |
| 4 | Llama 3.1 405B Instruct FP8 | 77.23 |
| 5 | GPT-4o (2024-08-06) | 75.15 |
| 6 | DeepSeek V3 (0324) | 73.46 |
| 7 | Llama 3.3 70B Instruct | 72.86 |
| 8 | Mistral-Small-Instruct-2409 | 72.85 |
| 9 | Llama 3.1 70B Instruct | 72.53 |
| 10 | Qwen 2 72B Instruct | 72.07 |
| 11 | Llama 3 70B Instruct | 71.21 |
| 12 | Qwen 2.5 32B Instruct | 71.15 |
| 13 | GPT-4o Mini (2024-07-18) | 71.15 |
| 14 | Bielik-11B-v2.3-Instruct | 70.86 |
| 15 | Mistral Small 3 | 70.52 |
Interactive version: theaggregate.ai/benchmark?slug=polish-eq-bench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.