Methodology

These benchmarks ensure fair, reproducible results. Run them yourself to verify!

Setup

• Environment: Node.js v22.14.0, mid-tier CPU (e.g., Intel i7), 16GB RAM.
• Library Versions: Typyn v1.0.1, Zod v4.1.12.
• Tool: Tinybench (~1M iterations per test, min 50 samples, RME <2%).
• Data: Valid inputs with realistic chains (e.g., v.string().min(5).email().parse('test@example.com')).
• Runs: Averaged from 2 executions; focused on .parse() (strict mode).

Test Cases

• Primitives: Simple type + 2-3 chains.
• Composites: 5-10 items/keys with nested primitives.
• Advanced: Standard formats (e.g., valid UUID string).

Running Locally

1. Clone repo: git clone https://github.com/ManojKumar2920/typyn.
2. Install: pnpm install.
3. Run: pnpm bench (outputs tables + CSV).
4. Compare: Add Zod to benchmarks/index.ts for side-by-side.

Notes: Results vary by hardware (~10-20%); use for relative perf. No invalid data (focus on happy path speed).

Back to Overview | Full Results

---

Last updated: October 23, 2025.