LibEvoBench

What is LibEvoBench?

LibEvoBench is a multi-task benchmark for evaluating code generation models on API evolution across versions of Python libraries. It includes tasks spanning…

Released

2026-06-24

Evaluates

Code, Code & Software, Code generation, Factuality, Software & AI Compute

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2606.25402

LibEvoBench paper

• Paper https://arxiv.org/abs/2606.25402

Which sources cite LibEvoBench?

• LibEvoBench

Related Code, Code & Software, Code generation, Factuality, Software & AI Compute benchmarks

• Active-SWE
• AFT-Bench
• Agent App Benchmark
• Agent Retrieval Bench
• Agent Systems Benchmark (ASB)
• Agentic Evaluation in 2D Mazes
• Agentic Security Benchmark
• AgentInfraBench