← all datasets

MEnvBench

Emerging
2papers using it
59HF downloads
3HF likes
2026first seen

MEnvBench: Multi-Language Environment Construction Benchmark πŸ“‹ Dataset Description MEnvBench is a comprehensive benchmark for evaluating multi-language environment building and test execution capabilities, comprising 1,000 task instances (10 languages Γ— 20 repositories Γ— 5 instances) selected from 200 high-quality ope

Papers using MEnvBench (1)

MEnvBench β€” datasets β€” ai-agents