Individual experiment result
ScaLA (DE)
Mistral Large 4 (preview) Mistral
Matthews correlation coefficient reasoning off provider-internal
61.3%
Matthews correlation coefficient
Run details
- Date
- 2026-10-06
- Test cases
- 2,048
- Median latency
- 0.66s
- Total cost
- $1.12
- Avg. prompt tokens
- 799.4
- Avg. answer tokens
- 2.7
- Avg. reasoning tokens
- —
- 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
- mistralai/mistral-large-4-0
- Run ID
2026-10-06T17-24-44__openrouter__mistralai__mistral-large-4-0__scala_de