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 (parseability-adjusted). Source: huggingface.co. Saturation forecast: Estimated already saturated. 102 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Large 2 (Jul) | 78.07 | #484 |
| 2 | Mistral Large 2 (Nov) Instruct (2411) | 77.29 | #465 |
| 3 | Llama 3.1 405B Instruct FP8 | 77.23 | #389 |
| 4 | GPT-4o (2024-08-06) | 75.15 | #336 |
| 5 | GPT-4 Turbo | 74.59 | #404 |
| 6 | DeepSeek V3 (0324) | 73.46 | #340 |
| 7 | Mistral-Small-Instruct-2409 | 72.85 | #702 |
| 8 | Llama-PLLuM-70B-chat | 72.56 | #759 |
| 9 | Llama 3.1 70B Instruct | 72.53 | #561 |
| 10 | Bielik-11B-v2.5-Instruct | 72 | #567 |
| 11 | Qwen 2 72B Instruct | 71.23 | #576 |
| 12 | Llama 3 70B Instruct | 71.21 | #634 |
| 13 | GPT-4o Mini (2024-07-18) | 71.15 | #558 |
| 14 | Qwen 2.5 32B Instruct | 71.15 | #498 |
| 15 | Bielik-11B-v2.3-Instruct | 70.86 | #693 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=polish-eq-bench · How It Works · Data refreshed daily, snapshot 2026-10-20.