German Language LLM Index
a PeerBench project

Individual experiment result

SB10K (DE)

Nemotron 3.5 Bolt 74B-A9B (IQ2) NVIDIA

Matthews correlation coefficient reasoning off iq2
52.8%
Matthews correlation coefficient

Run details

Date
2026-10-15
Test cases
1,024
Median latency
16.48s
Total cost
—
Avg. prompt tokens
859.3
Avg. answer tokens
2.6
Avg. reasoning tokens
—
Quantization
iq2

About this benchmark

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

3-class sentiment native · Native German Source ↗

Provenance

Run ID
2026-10-15T20-40-39__openai__nvidia__NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB__sb10k_de