German Language LLM Index
a PeerBench project

Individual experiment result

ScaLA (DE)

Qwen3.8 27B (BF16) Alibaba

Matthews correlation coefficient reasoning off bf16
66.9%
Matthews correlation coefficient

Run details

Date
2026-08-15
Test cases
2,048
Median latency
2.13s
Total cost
Avg. prompt tokens
976.9
Avg. answer tokens
2.6
Avg. reasoning tokens
Quantization
bf16

About this benchmark

Native German linguistic acceptability — is the sentence grammatical (ja/nein)? generate_until; primary metric MCC (accuracy recorded as secondary).

Binary acceptability native · Native German Source ↗

Provenance

Canonical model
qwen/qwen3.8-27b
Run ID
2026-08-14T23-45-10__openai__Qwen__Qwen3.8-27B__scala_de