optimizing-performance

Speed up Zod validation by using safeParse to avoid exceptions, defining reusable schemas at module scope, validating arrays in bulk, and employing discriminated unions. These patterns reduce redundant parsing overhead and improve throughput in Zod v4 workloads.

Use safeParse when handling potentially invalid data without crashing your application. It returns a result object with success status instead of throwing exceptions, making it safer for API requests and user input validation while maintaining performance.

Validate large arrays by defining schemas at module level for reuse, applying bulk validation patterns, and using discriminated unions where appropriate. This approach avoids schema recreation per request and speeds up Zod v4 validation of array-heavy workloads.

Use passthrough to preserve extra object keys instead of stripping them, avoiding unnecessary processing. This optimization is valuable when you need to retain unmapped data in API validation or bulk validation scenarios without performance penalties.

Module-level schemas are parsed once and reused across requests, eliminating repeated schema construction overhead. This pattern is essential for high-throughput applications performing internal validation, API request validation, and bulk array validation efficiently.

No. Order refinements strategically—apply cheapest validations first, then expensive ones. Use discriminated unions and passthrough patterns instead of refinements where possible to maintain performance in Zod v4 schemas.