German Language LLM Index
a PeerBench project

Individual experiment result

GermEval NER (DE)

Mistral Large 4 (preview) Mistral

Micro-F1 excluding MISC reasoning off provider-internal
79.9%
Micro-F1 excluding MISC

Run details

Date
2026-10-06
Test cases
1,024
Median latency
1.03s
Total cost
$0.82
Avg. prompt tokens
1,070.1
Avg. answer tokens
35
Avg. reasoning tokens
—
Quantization
provider-internal

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

Run ID
2026-10-06T17-29-38__openrouter__mistralai__mistral-large-4-0__germeval_de