German Language LLM Index
a PeerBench project

Individual experiment result

GermEval NER (DE)

Qwen3.6 27B (BF16) Alibaba

Micro-F1 excluding MISC reasoning off bf16
82.7%
Micro-F1 excluding MISC

Run details

Date
2026-08-15
Test cases
1,024
Median latency
2.79s
Total cost
Avg. prompt tokens
1,154.3
Avg. answer tokens
25.8
Avg. reasoning tokens
Quantization
bf16

About this benchmark

Native German named-entity recognition (person/location/org/misc) emitted as JSON; seqeval micro-F1 excluding MISC.

Named-entity recognition native · Native German Source ↗

Provenance

Canonical model
qwen/qwen3.6-27b
Run ID
2026-08-15T08-50-25__openai__Qwen__Qwen3.6-27B__germeval_de