Evaluates LLMs on Polish linguistic and cultural knowledge across 6 categories: art & entertainment, culture & tradition, geography, grammar, history, and vocabulary. Accuracy (0-100) per category. Created by Dadas et al. (2025).
Geography is the reported evaluation metric for PLCC. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.
Higher is better
Muted rows were not state of the art when published — an earlier or same-year result already scored better.
Culture And Tradition is the reported evaluation metric for PLCC. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.
Higher is better
Muted rows were not state of the art when published — an earlier or same-year result already scored better.
History is the reported evaluation metric for PLCC. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.
Higher is better
Muted rows were not state of the art when published — an earlier or same-year result already scored better.
Average is the reported evaluation metric for PLCC. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.
Higher is better
Muted rows were not state of the art when published — an earlier or same-year result already scored better.
| Rank | Model | Trust | Score | Year | Links | Fix |
|---|---|---|---|---|---|---|
| 01 | Gemini-3.1-Pro-Preview | verified | 97 | 2026 | Source ↗ | Looks wrong? |
| 02 | Gemini-3.0-Pro-Preview | verified | 95.833333 | 2026 | Source ↗ | Looks wrong? |
| 03 | GPT-5.4-2026-03-05 (high reasoning) | verified | 92.166667 | 2026 | Source ↗ | Looks wrong? |
| 04 | Gemini-2.5-Pro-Preview-06-05 | verified | 92.166667 | 2026 | Source ↗ | Looks wrong? |
| 05 | Gemini-3-Flash-Preview | verified | 91.666667 | 2026 | Source ↗ | Looks wrong? |
| 06 | GPT-5-Pro-2025-10-06 (high reasoning) | verified | 91 | 2026 | Source ↗ | Looks wrong? |
| 07 | Grok 4 | verified | 90.5 | 2026 | Source ↗ | Looks wrong? |
| 08 | GPT-5.4-2026-03-05 (low reasoning) | verified | 90.5 | 2026 | Source ↗ | Looks wrong? |
| 09 | GPT-5-2025-08-07 | verified | 89.5 | 2026 | Source ↗ | Looks wrong? |
| 10 | Gemini-2.5-Pro-Exp-03-25 | verified | 89.5 | 2026 | Source ↗ | Looks wrong? |
| 11 | GPT-5.2-2025-12-11 (xhigh reasoning) | verified | 89.333333 | 2026 | Source ↗ | Looks wrong? |
| 12 | O3-2025-04-16 | verified | 89.166667 | 2026 | Source ↗ | Looks wrong? |
| 13 | O1-2024-12-17 | verified | 89.166667 | 2026 | Source ↗ | Looks wrong? |
Vocabulary is the reported evaluation metric for PLCC. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.
Higher is better
Muted rows were not state of the art when published — an earlier or same-year result already scored better.
| Rank | Model | Trust | Score | Year | Links | Fix |
|---|---|---|---|---|---|---|
| 01 | Gemini-3.1-Pro-Preview | verified | 96 | 2026 | Source ↗ | Looks wrong? |
| 02 | Gemini-3.0-Pro-Preview | verified | 95 | 2026 | Source ↗ | Looks wrong? |
| 03 | GPT-5-Pro-2025-10-06 (high reasoning) | verified | 92 | 2026 | Source ↗ | Looks wrong? |
| 04 | GPT-5-2025-08-07 | verified | 91 | 2026 | Source ↗ | Looks wrong? |
| 05 | GPT-5.4-2026-03-05 (high reasoning) | verified | 91 | 2026 | Source ↗ | Looks wrong? |
| 06 | Gemini-2.5-Pro-Exp-03-25 | verified | 90 | 2026 | Source ↗ | Looks wrong? |
| 07 | Gemini-2.5-Pro-Preview-06-05 | verified | 90 | 2026 | Source ↗ | Looks wrong? |
| 08 | GPT-5.1-2025-11-13 (high reasoning) | verified | 90 | 2026 | Source ↗ | Looks wrong? |
| 09 | O3-2025-04-16 | verified | 90 | 2026 | Source ↗ | Looks wrong? |
Art And Entertainment is the reported evaluation metric for PLCC. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.
Higher is better
Muted rows were not state of the art when published — an earlier or same-year result already scored better.
| Rank | Model | Trust | Score | Year | Links | Fix |
|---|---|---|---|---|---|---|
| 01 | Gemini-3.1-Pro-Preview | verified | 95 | 2026 | Source ↗ | Looks wrong? |
| 02 | Gemini-3.0-Pro-Preview | verified | 95 | 2026 | Source ↗ | Looks wrong? |
| 03 | GPT-5.4-2026-03-05 (high reasoning) | verified | 91 | 2026 | Source ↗ | Looks wrong? |
| 04 | Gemini-3-Flash-Preview | verified | 91 | 2026 | Source ↗ | Looks wrong? |
| 05 | Gemini-2.5-Pro-Preview-06-05 | verified | 91 | 2026 | Source ↗ | Looks wrong? |
| 06 | GPT-4.5-preview-2025-02-27 | verified | 90 | 2026 | Source ↗ | Looks wrong? |
Grammar is the reported evaluation metric for PLCC. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.
Higher is better
Muted rows were not state of the art when published — an earlier or same-year result already scored better.
| Rank | Model | Trust | Score | Year | Links | Fix |
|---|---|---|---|---|---|---|
| 01 | Gemini-3.1-Pro-Preview | verified | 93 | 2026 | Source ↗ | Looks wrong? |
| 02 | Gemini-3.0-Pro-Preview | verified | 91 | 2026 | Source ↗ | Looks wrong? |
| 03 | GPT-5.4-2026-03-05 (high reasoning) | verified | 90 | 2026 | Source ↗ | Looks wrong? |
| 04 | Grok 4 | verified | 90 | 2026 | Source ↗ | Looks wrong? |
| 05 | GPT-5.2-2025-12-11 (xhigh reasoning) | verified | 89 | 2026 | Source ↗ | Looks wrong? |