Individual experiment result
MMLU-ProX (DE)
Tencent HY3-Preview Tencent
Exact match reasoning off provider-internal
73.2%
Exact match
Run details
- Date
- 2026-05-27
- Test cases
- 11,759
- Median latency
- 6.14s
- Total cost
- $2.76
- Avg. prompt tokens
- 1,910.2
- Avg. answer tokens
- 586.3
- Avg. reasoning tokens
- —
- Quantization
- provider-internal
About this benchmark
German MMLU-ProX across 14 subjects; primary metric accuracy. Legacy rows have 11,759 items. Corrected v1.1 rows have 11,737 after a frozen, model-independent manifest excludes 20 duplicated-gold translation collisions, one independently reviewed ambiguous source item, and one independently confirmed notation-translation collapse; the row's item count identifies the definition.
Subject breakdown
| Subject | Score |
|---|---|
| biology | 85.8% |
| business | 78.3% |
| chemistry | 75.6% |
| computer science | 76.8% |
| economics | 82.3% |
| engineering | 64.4% |
| health | 73.7% |
| history | 71.4% |
| law | 45.8% |
| math | 86.5% |
| other | 72.1% |
| philosophy | 69.1% |
| physics | 64.7% |
| psychology | 80.5% |
Provenance
- Canonical model
- tencent/hy3-preview
- Run ID
2026-05-27T23-43-24__openrouter__tencent__hy3-preview