Benchmark RadarMulti-LCB
Multi-LCB evaluates code generation across twelve programming languages (C++, C#, Python, Java, Rust, Go, TypeScript, JavaScript, Ruby, Kotlin, Scala, PHP) by transforming Python tasks from LiveCodeBench while preserving its contamination controls and evaluation protocol.
- Released
- 2026-06-18
- Readiness
- Runnable
- Primary field
- General AI
Why it matters
LiveCodeBench restricted code evaluation to Python; Multi-LCB addresses the gap by enabling cross-language assessment, revealing Python overfitting and language-specific contamination in LLMs, and supporting robust multilingual code evaluation for real-world software engineering.
Motivation
LiveCodeBench (LCB) has recently become a widely adopted benchmark for evaluating large language models (LLMs) on code-generation tasks.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.