Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
RepoBenchgithub.com · Low evidenceRepoBench is a benchmark for evaluating repository-level code auto-completion systems, accepted at ICLR 2024. It provides datasets and evaluation scripts for Python and Java, with settings for cross-file and in-file completion tasks.
Parse Score