Sourcegraph Blog
2026-03-03 00:00 UTC
By Stephanie Jarmak
USR-0064-20260303-ai-specialis-c3719a92
CodeScaleBench: Testing coding agents on large codebases and multi-repo software engineering tasks
The initial findings from CodeScaleBench, a new benchmark designed to evaluate coding agents against the true complexity of enterprise software development, including large codebases and multi-repository tasks.
The initial findings from CodeScaleBench, a new benchmark designed to evaluate coding agents against the true complexity of enterprise software development, including large codebases and multi-repository tasks.
Full article content could not be extracted automatically. Read the original below.