JavaScript Code Evals allow you to write custom Code Evals in JavaScript in addition to Python. This gives you the flexibility to choose the language that best fits your workflow and expertise when building custom evaluations for your LLM applications.

Why use JS Code Evals:

Speed: Early testing shows JavaScript evals run approximately 15x faster than Python equivalents, dramatically reducing evaluation time for your projects.

Flexibility: Use the language you're most comfortable with or that best matches your tech stack. If you're already working in a JavaScript/TypeScript environment, you can now stay in that ecosystem end-to-end.

Performance at Scale: Faster execution means quicker iteration cycles, especially valuable when running large evaluation suites or testing multiple prompt variations.

See the docs: https://docs.langchain.com/langsmith/online-evaluations#configure-a-custom-code-evaluator