Individual experiment result
ScaLA (DE)
Gemini 3.8 Flash Google
Matthews correlation coefficient reasoning on / locked provider-internal
80.6%
Matthews correlation coefficient
Run details
- Date
- 2026-09-04
- Test cases
- 2,048
- Median latency
- 3.45s
- Total cost
- $4.70
- Avg. prompt tokens
- 777.7
- Avg. answer tokens
- 1.5
- Avg. reasoning tokens
- 148.9
- Quantization
- provider-internal
About this benchmark
Native German linguistic acceptability — is the sentence grammatical (ja/nein)? generate_until; primary metric MCC (accuracy recorded as secondary).
Provenance
- Canonical model
- google/gemini-3.8-flash
- Run ID
2026-09-04T15-05-05__vertex_ai__gemini-3.8-flash__scala_de