Benchmark RadarAI BENCHMARK PROFILE
SolidityBench
SolidityBench contains 5,470 repository-level Solidity smart contracts with natural language descriptions, plus SolidityScore, a semantic metric emphasizing domain-critical constructs. It evaluates code generation models.
- Released
- 2026-06-18
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
Domain-specific code generation lacks benchmarks. SolidityBench provides a resource to measure structural and semantic correctness in high-stakes smart contracts.
Motivation
Large Language Models (LLMs) have shown strong capabilities in general-purpose code generation, but their effectiveness in specialized software domains remains underexplored.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.