German Language LLM Index
a PeerBench project

Individual experiment result

SB10K (DE)

Muse Spark Open Meta

Matthews correlation coefficient reasoning on / locked provider-internal
66.4%
Matthews correlation coefficient

Run details

Date
2026-10-16
Test cases
1,024
Median latency
1.48s
Total cost
$2.77
Avg. prompt tokens
871.9
Avg. answer tokens
6.5
Avg. reasoning tokens
404.1
Quantization
provider-internal

About this benchmark

Native German Twitter sentiment (positive/neutral/negative), generate_until variant; primary metric MCC.

3-class sentiment native · Native German Source ↗

Provenance

Canonical model
meta/muse-spark-open
Run ID
2026-10-16T11-38-00__openrouter__meta__muse-spark-open__sb10k_de