Valibot vs Zod — Can a 1KB Validator Replace Zod?

Modern JavaScript applications rely heavily on external data sources. APIs, forms, query parameters, cookies, and configuration files all deliver data that your application must trust and process.

But in practice, external data cannot be trusted.

Fields disappear. Types change. Backends evolve. A simple typo in a property name can break an entire page.

Schema validation libraries solve this problem by adding a runtime validation layer between your application and external data.

Two libraries dominate this space today:

Zod
Valibot

Zod has been the standard for years. Valibot, however, is a newer library designed with a different philosophy: smaller bundles and faster validation.

Why Validation Libraries Matter

TypeScript provides compile‑time guarantees, but it cannot guarantee runtime correctness.

Example:

const response = await fetch("/api/user/1")
const user = await response.json()

Runtime data might not match expectations.

Validation libraries ensure data integrity before it reaches business logic.

Validating API Responses with Valibot

import * as v from "valibot"

const ProductSchema = v.object({
  id: v.number(),
  title: v.string(),
  price: v.number(),
  rating: v.number(),
  images: v.array(v.string())
})

type Product = v.InferOutput<typeof ProductSchema>

export async function loadProduct(id:number):Promise<Product>{

  const res = await fetch(`https://dummyjson.com/products/${id}`)
  const data = await res.json()

  try{
    return v.parse(ProductSchema,data)
  }
  catch(error){
    console.error("Invalid API response",error)
    throw new Error("Product validation failed")
  }

}

Validating Form Input

import * as v from "valibot"

const RegistrationSchema = v.object({

  username:v.pipe(
    v.string(),
    v.minLength(3)
  ),

  email:v.pipe(
    v.string(),
    v.email()
  ),

  age:v.pipe(
    v.string(),
    v.digits(),
    v.transform(Number)
  ),

  password:v.pipe(
    v.string(),
    v.minLength(6)
  )

})

type RegistrationData = v.InferOutput<typeof RegistrationSchema>

Cross‑Field Validation

const RegistrationSchema = v.pipe(

  v.object({

    email:v.pipe(
      v.string("Email must be text"),
      v.email("Enter a valid email")
    ),

    age:v.pipe(
      v.string(),
      v.digits("Age must contain digits"),
      v.transform(Number),
      v.minValue(18,"You must be at least 18")
    ),

    password:v.pipe(
      v.string(),
      v.minLength(6,"Password must contain at least 6 characters")
    ),

    confirmPassword:v.string()

  }),

  v.forward(
    v.partialCheck(
      [["password"],["confirmPassword"]],
      ({password,confirmPassword}) => password === confirmPassword,
      "Passwords do not match"
    ),
    ["confirmPassword"]
  )

)

Code Comparison

Valibot

import * as v from "valibot"

const BookSchema = v.object({
  title:v.string(),
  price:v.number()
})

v.parse(BookSchema,{title:"JavaScript Guide",price:30})

Zod

import {z} from "zod"

const BookSchema = z.object({
  title:z.string(),
  price:z.number()
})

BookSchema.parse({title:"JavaScript Guide",price:30})

Benchmark Scripts

Valibot

import * as v from "valibot"

const schema = v.object({
 id:v.number(),
 name:v.string(),
 price:v.number(),
 rating:v.number(),
 tags:v.array(v.string())
})

const sample = {
 id:1,
 name:"Laptop",
 price:1500,
 rating:4.8,
 tags:["tech","computer"]
}

function benchmarkValibot(iterations:number){

 const start = performance.now()

 for(let i=0;i<iterations;i++){
  v.parse(schema,sample)
 }

 return performance.now() - start

}

console.log("Valibot:",benchmarkValibot(10000),"ms")

Zod

import {z} from "zod"

const schema = z.object({
 id:z.number(),
 name:z.string(),
 price:z.number(),
 rating:z.number(),
 tags:z.array(z.string())
})

const sample = {
 id:1,
 name:"Laptop",
 price:1500,
 rating:4.8,
 tags:["tech","computer"]
}

function benchmarkZod(iterations:number){

 const start = performance.now()

 for(let i=0;i<iterations;i++){
  schema.parse(sample)
 }

 return performance.now() - start

}

console.log("Zod:",benchmarkZod(10000),"ms")

Benchmark Results

Valid data

Library	min (ms)	max (ms)
Valibot	26.58	27.31
Zod	67.02	70.25

Invalid data

Library	min (ms)	max (ms)
Valibot	48.63	54.70
Zod	143.63	147.79

Bundle Size

Replacing Zod with Valibot reduced bundle size by ~11.4 KB (gzip) in a medium project.

Popularity

Library	Weekly downloads
Valibot	~1.2M
Zod	~89.5M

Conclusion

Validation libraries protect applications from unreliable external data.

Valibot proves that validation can be fast, lightweight, and developer‑friendly.

For new projects, Valibot is definitely worth considering.