Individual experiment result
MuSR (DE)
Gemini 4 Argon Google
Accuracy reasoning on / locked provider-internal
93.5%
Accuracy
Run details
- Date
- 2026-10-13
- Test cases
- 553
- Median latency
- 9.05s
- Total cost
- $7.62
- Avg. prompt tokens
- 1,493.2
- Avg. answer tokens
- 709.3
- Avg. reasoning tokens
- 369.5
- 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 | 95.6% |
| object placements | 89.1% |
| team allocation | 92.5% |
Provenance
- Canonical model
- google/gemini-4-argon
- Run ID
2026-10-13T19-10-37__vertex_ai__gemini-4-argon__musr_de