86.1%
Accuracy
Run details
- Date
- 2026-08-15
- Test cases
- 553
- Median latency
- 40.69s
- Total cost
- —
- Avg. prompt tokens
- 1,489.1
- Avg. answer tokens
- 1,772.5
- Avg. reasoning tokens
- —
- Quantization
- bf16
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 | 88.4% |
| object placements | 75.0% |
| team allocation | 86.6% |
Provenance
- Canonical model
- qwen/qwen3.8-27b
- Run ID
2026-08-14T23-10-23__openai__Qwen__Qwen3.8-27B__musr_de