Open PL LLM - PolEmo2-OUT (multiple choice, 5-shot): leaderboard
Sentiment classification of Polish online reviews from the out-of-domain PolEmo 2.0 test set, whose review domains are absent from training. Open PL LLM Leaderboard (SpeakLeash), 5-shot, multiple choice variant: the options are ranked by log-likelihood; accuracy (%), higher is better. Runs not available on the Hugging Face Hub are excluded.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around February 2027. 320 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Large 2 (Jul) | 83.4 | #490 |
| 2 | Yi 1.5 34B Chat | 82.79 | #798 |
| 3 | Qwen 2.5 14B Instruct | 82.39 | #654 |
| 4 | Qwen 2 72B | 82.19 | #515 |
| 5 | EuroLLM-22B-Instruct-2512 | 81.98 | #704 |
| 6 | Qwen 2 72B Instruct | 81.78 | #579 |
| 7 | Qwen 2.5 72B | 81.38 | #512 |
| 8 | Mistral Large 2 (Nov) Instruct (2411) | 80.97 | #480 |
| 9 | Llama 4 Scout Instruct | 80.36 | #547 |
| 10 | Qwen 2.5 72B Instruct | 80.16 | #469 |
| 11 | Qwen 2.5 32B Instruct | 79.96 | #509 |
| 12 | cogito-v1-preview-qwen-14B | 79.76 | |
| 13 | Qwen 2.5 14B | 79.76 | #716 |
| 14 | Bielik-11B-v2.0-Instruct | 79.15 | #750 |
| 15 | SOLAR-10.7B-Instruct-v1.0 | 78.95 | #1066 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-pl-llm-polemo2-out-multiple-choice-5-shot · How It Works · Data refreshed daily, snapshot 2026-11-05.