Individual experiment result
SB10K (DE)
Claude Fable 5 Anthropic
Matthews correlation coefficient reasoning on / locked provider-internal
67.3%
Matthews correlation coefficient
Run details
- Date
- 2026-07-06
- Test cases
- 1,024
- Median latency
- 4.60s
- Total cost
- $18.05
- Avg. prompt tokens
- 1,352.6
- Avg. answer tokens
- 19.3
- Avg. reasoning tokens
- —
- Quantization
- provider-internal
About this benchmark
Native German Twitter sentiment (positive/neutral/negative), generate_until variant; primary metric MCC.
Provenance
- Canonical model
- anthropic/claude-fable-5
- Run ID
2026-07-06T09-59-46__openai__cc__claude-fable-5__sb10k_de