Open PL LLM - PSC (multiple choice, 5-shot): leaderboard
Polish Summaries Corpus: deciding whether a summary belongs to a news article. Open PL LLM Leaderboard (SpeakLeash), 5-shot, multiple choice variant: the options are ranked by log-likelihood; binary F1 (%), higher is better. Runs not available on the Hugging Face Hub are excluded.
Metric: Binary F1 (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 320 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 72B | 97.25 | #512 |
| 2 | Qwen 2.5 14B | 96.48 | #716 |
| 3 | Qwen 2.5 32B | 94.59 | #544 |
| 4 | Bielik-11B-v2.0-Instruct | 94.45 | #750 |
| 5 | Qwen 1.5 72B | 94.27 | #752 |
| 6 | QwQ 32B-Preview | 93.4 | #671 |
| 7 | Mistral-Small-Instruct-2409 | 92.63 | #709 |
| 8 | Bielik-11B-v2.3-Instruct | 92.41 | #700 |
| 9 | Qwen 1.5 72B Chat | 92.36 | #782 |
| 10 | Bielik-11B-v2.2-Instruct | 91.68 | #669 |
| 11 | QwQ-32B | 91.23 | #422 |
| 12 | Bielik-11B-v2.1-Instruct | 90.43 | #679 |
| 13 | Llama 3 70B Instruct | 90.38 | #640 |
| 14 | EuroLLM-22B-Instruct-2512 | 90.22 | #704 |
| 15 | c4ai-command-r-v01 | 89.84 | #993 |
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-psc-multiple-choice-5-shot · How It Works · Data refreshed daily, snapshot 2026-11-05.