German Language LLM Index
a PeerBench project

Individual experiment result

ScaLA (DE)

GLM-5.3 Flash Z.ai

Matthews correlation coefficient reasoning on / locked fp8
77.2%
Matthews correlation coefficient

Run details

Date
2026-08-30
Test cases
2,048
Median latency
2.35s
Total cost
Avg. prompt tokens
920.6
Avg. answer tokens
3.8
Avg. reasoning tokens
240.9
Quantization
fp8

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
z-ai/glm-5.3-flash
Run ID
2026-08-30T16-35-15__openrouter__z-ai__glm-5.3-flash__scala_de