German LLM Index - INCLUDE v1.2: leaderboard
German Language LLM Index (dach.peerbench.ai, a PeerBench project, 2026), which reruns every model itself with lm-evaluation-harness: chat template, temperature 0, up to 8,192 generated tokens, thinking models set to minimal reasoning effort where the API allows it and other models run as served. Labels carry (Reasoning off) when the site classed the run as reasoning-off (at most ten reasoning tokens, or under 0.1% of its completion tokens); other rows reasoned at the provider default or at the effort the label names. German part of INCLUDE (CohereLabs include-base-44): native German exam and licensing questions on regional knowledge such as history, law and civics, four options, in the audited v1.2 set of 107 valid questions (32 defective source questions removed). Exact match with strict final-answer parsing. Higher is better. Source: https://dach.peerbench.ai/
Metric: Exact match, strict final-answer parsing (%). Source: dach.peerbench.ai. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.7 Flash | 93.46 | #24 |
| 2 | Gemini 4 Argon | 89.72 | #9 |
| 3 | GPT-5.6 Terra (Non-reasoning) | 88.79 | #40 (GPT-5.6 Terra) |
| 4 | DeepSeek V4 Pro (0813) (Non-reasoning) | 87.85 | #58 (DeepSeek V4 Pro (0813)) |
| 5 | GLM-5.3 Flash | 86.92 | #77 |
| 6 | Muse Spark Open | 86.92 | #42 |
| 7 | Muse Glimmer 30B | 85.98 | #191 |
| 8 | GPT-5.6 Luna (Non-reasoning) | 85.98 | #79 (GPT-5.6 Luna) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=german-llm-index-include-v1-2 · How It Works · Data refreshed daily, snapshot 2026-11-05.