German Language LLM Index
a PeerBench project

Individual experiment result

ScaLA (DE)

Claude Fable 5 Anthropic

Matthews correlation coefficient reasoning on / locked provider-internal
83.6%
Matthews correlation coefficient

Run details

Date
2026-07-06
Test cases
2,048
Median latency
5.20s
Total cost
$44.77
Avg. prompt tokens
1,661.5
Avg. answer tokens
28
Avg. reasoning tokens
Quantization
provider-internal

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

Run ID
2026-07-06T11-27-02__openai__cc__claude-fable-5__scala_de