MCR-Bench
Explore score history ↗MCR-Bench is a benchmark for evaluating reproducibility in mission-critical LLM tasks, measuring output consistency across heterogeneous hardware.
- Released
- 2026-06-19
- Institutions
- University of California, Berkeley · University of Chinese Academy of Sciences
- Evaluates
- Finance, Language & Knowledge, cs.LG
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
MCR-Bench paper
No reported scores are on record for this benchmark yet.