MOEVE - Party Positions: leaderboard
MOEVE (Modelle fuer die Oeffentliche Verwaltung Evaluieren, Bundesdruckerei, 2026) benchmarks language models for the German public sector with German prompts and German datasets, many of them from public administration. This board is accuracy in percent at classifying the positions of German political parties on 4,788 Wahl-O-Mat statements; higher is better.
Metric: Accuracy (%) at classifying the positions of German political parties on 4,788 Wahl-O-Mat statements; German prompts, default decoding, open-weight models on vLLM or Ollama at the quantization listed in the paper's model table, others by API; the paper prints only the top 10 of 39 evaluated models; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o Mini | 67.1 | #1623 |
| 2 | DeepSeek R1 | 67.1 | #473 |
| 3 | GPT-OSS-120B | 66.7 | #700 |
| 4 | Llama 3.3 70B | 66.3 | #1498 |
| 5 | Mistral Small 3.1 | 64.5 | #1689 |
| 6 | GPT-4o | 62.4 | #741 |
| 7 | GPT-OSS-20B | 61.9 | #1147 |
| 8 | Mistral Large 3 | 61 | #817 |
| 9 | DeepSeek R1 Distill Qwen 32B | 60.1 | #1788 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=moeve-party-positions · How It Works · Data refreshed daily, snapshot 2026-11-13.