Individual experiment result
MuSR (DE)
Ling 3.1 Flash inclusionAI
Accuracy reasoning off provider-internal
84.6%
Accuracy
Run details
- Date
- 2026-10-16
- Test cases
- 553
- Median latency
- 6.50s
- Total cost
- $0.07
- Avg. prompt tokens
- 1,791.3
- Avg. answer tokens
- 598.5
- Avg. reasoning tokens
- —
- Quantization
- provider-internal
About this benchmark
German MuSR v1.1 defect-filtered benchmark: 553 questions (11 truncated team-allocation translations excluded), generate_until cot+; primary metric accuracy.
Subject breakdown
| Subject | Score |
|---|---|
| murder mystery | 81.6% |
| object placements | 78.1% |
| team allocation | 89.5% |
Provenance
- Canonical model
- inclusionai/ling-3.1-flash
- Run ID
2026-10-16T12-57-02__openrouter__inclusionai__ling-3.1-flash__musr_de