Individual experiment result
GermEval NER (DE)
Gemini 3 Flash Preview Google
Micro-F1 excluding MISC reasoning off provider-internal
84.7%
Micro-F1 excluding MISC
Run details
- Date
- 2026-06-25
- Test cases
- 1,024
- Median latency
- 1.20s
- Total cost
- $0.62
- Avg. prompt tokens
- 1,060
- Avg. answer tokens
- 24.9
- 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.
Provenance
- Canonical model
- google/gemini-3-flash-preview
- Run ID
2026-06-25T11-34-32__vertex_ai__gemini-3-flash-preview__germeval_de