Open PL LLM - PolEmo2-IN (generative, 5-shot): leaderboard
Sentiment classification of Polish online reviews from the in-domain PolEmo 2.0 test set. Open PL LLM Leaderboard (SpeakLeash), 5-shot, generative variant: the answer is extracted from generated text; accuracy (%), higher is better. Runs not available on the Hugging Face Hub are excluded.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 321 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 4 Scout Instruct | 89.34 | #547 |
| 2 | Qwen 2.5 14B Instruct | 89.06 | #654 |
| 3 | Mistral Large 2 (Nov) Instruct (2411) | 89.06 | #480 |
| 4 | Llama 3.1 405B Instruct FP8 | 89.06 | #400 |
| 5 | Mistral Large 2 (Jul) | 88.5 | #490 |
| 6 | DeepSeek V3 (0324) | 88.5 | #349 |
| 7 | Qwen 3 8B | 88.09 | #678 |
| 8 | cogito-v1-preview-qwen-14B | 88.09 | |
| 9 | Mistral Small 3 | 87.67 | #630 |
| 10 | Llama 3.1 70B Instruct | 87.53 | #567 |
| 11 | Qwen 2 72B Instruct | 87.26 | #579 |
| 12 | Llama 3.3 70B Instruct | 87.12 | #541 |
| 13 | Qwen 3 14B | 86.84 | #542 |
| 14 | Llama 3.1 Nemotron 70B Instruct | 86.84 | #563 |
| 15 | Qwen 2.5 72B Instruct | 86.57 | #469 |
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-in-generative-5-shot · How It Works · Data refreshed daily, snapshot 2026-11-05.