Individual experiment result
MuSR (DE)
Gemini 2.5 Flash Google
Accuracy reasoning off provider-internal
83.3%
Accuracy
Run details
- Date
- 2026-06-09
- Test cases
- 564
- Median latency
- 6.28s
- Total cost
- $1.77
- Avg. prompt tokens
- 1,464.9
- Avg. answer tokens
- 1,077.4
- 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 | 87.6% |
| object placements | 75.0% |
| team allocation | 81.2% |
Provenance
- Canonical model
- google/gemini-2.5-flash
- Run ID
2026-06-09T18-10-07__vertex_ai__gemini-2.5-flash__musr_de