Zod 4

https://zod.dev/v4

by bpierre • 1 year ago

782 255 1 year ago

zod.dev

90s_dev5• 1 year ago

TypeScript-first schema validation with static type inference

Is there a comparison guide? Never heard of this before, but I used io-ts and ajv.

cluckindan2• 1 year ago

Between GraphQL and Zod, there really is no comparison. The latter feels clunky and superfluous by design when the former ecosystem can smoothly provide static TS typings without extra work or schema duplication.

eknkc• 1 year ago

Also, I think teflon pans are much better than squids.

lyu07282• 1 year ago

Squid ink is black so are teflon pans, the latter feels clunky and superfluous by design when the former type of fruit can smoothly provide vitamins to vertically integrate with our stack.

gejose2• 1 year ago

Ah yes two totally different technologies - let's compare them like they're designed to solve the exact same problem

cluckindan2• 1 year ago

GraphQL is a different animal, but I was only comparing the features Zod implements to those found in the GraphQL ecosystem.

neilpa• 1 year ago

We're currently evaluating both Zod and ArkType as a replacement for earlier JSON schema validations. The thing I can't get over with Zod is the syntax is _so_ different from defining TS types.

Are there reasons to go with Zod over ArkType?

CharlieDigital• 1 year ago

You might like Typia better.

I recently put together a deck for this after some investigation: https://docs.google.com/presentation/d/1fToIKvR7dyvQS1AAtp4Y...

fkyoureadthedoc• 1 year ago

On the ArkType website they talk about speed and DX, but no mention of bundle size. I assume, and maybe this is an unfair assumption, that's because it's bad. I haven't come across ArkType until just now, but I was just checking out Valibot today because it's got a small bundle size.

rafram7• 1 year ago

https://bundlephobia.com/package/arktype@2.1.20: 38.3 KB

https://bundlephobia.com/package/zod@3.24.4: 64.3 KB

https://bundlephobia.com/package/valibot@1.1.0: 84.1 KB

petesergeant• 1 year ago

Everyone else is using Zod so the LLMs understand it well, there’s a big ecosystem, it’s unlikely to disappear overnight, and new hires may well already use it. It’s the boring solution.

satvikpendem6• 1 year ago

We also chose ArkType after evaluating Zod, Valibot, and Effect Schema, among others. It seemed to have the best developer experience while also being much faster. It also supports the Standard Schema project which I believe Effect Schema does not fully support.

bjacobso• 1 year ago

I believe Effect Schema fully supports Standard Schema, the issue is that it supports much more than Standard Schema, so not all schemas will work and thus provide compile time errors.

sesm• 1 year ago

Does ArkType support generating data from schema?

mapcars• 1 year ago

Could you elaborate on how exactly do you use these? I was assuming its for some kind of contract testing, but there are mentions of bundle size, is it used on fetching data to fail early if the format is wrong, something like that?

causal3• 1 year ago

Zod is a lot better than some of the alternative solutions I've seen. That said, the need for this sort of explicit validation always felt like a failure of where modern web development has taken us. It's so frustrating how full-stack development requires so many ways of describing the same shapes: JS input validation, Swagger for API definition, server-side input validation, ORM for schema conformance, and TypeScript often requiring separate definitions on server and client side. It's so tedious.

koolba• 1 year ago

But that’s the whole point of something like this. You do it once and dynamically generate everything else downstream. So change it once in the zod schema and it propagates with type checking through your entire app.

The zod schema becomes the source of truth.

worldsayshi2• 1 year ago

Does zod really support that? What if the code base start out with a go backend and a TS/JS component is only added later? It would be nice if the source of truth was a bit more language agnostic.

esafak2• 1 year ago

Does it? I don't use Zod on my JVM backend. How would I?

In addition to using OpenAPI, I generate TS interfaces from my data classes in a Gradle task.

kellengreen• 1 year ago

Does it?

Most Go, Java, and Python APIs are practically all Swagger based.

worldsayshi2• 1 year ago

Yes there should at least be a way to consolidate so that we can agree on format to treat as the source of truth that we can generate the other formats from.

crab_galaxy• 1 year ago

OpenAPI TypeScript is the closest thing I’ve found to perfection when your API is written in a different language than your client.

https://openapi-ts.dev/

satvikpendem6• 1 year ago

How else would you do it? You could use something like trpc for full end to end type safety but that requires using TypeScript on both the frontend and backend (which not everyone does) and also locking yourself into only the web platform (so no mobile, etc).

jauntywundrkind• 1 year ago

TypeScript's insistence on being a static checking/compile time only system is such a sad wound to the whole ecosystem. I don't want TypeScript to be a runtime checker, but I want it to export useful usable type data for classes, functions, objects. TypeScript feels like the best source of truth we have, but instead so many of the attempts to reflect on what we have are all going it alone, having to define their own model and their own builders for describing what stuff is. You've mentioned 5 major areas where reflection is a core part of the mission, and each of these have multiple implementations.

There is the old TypeScript type emitter, reflect-metadata, which would provide some type information at runtime, but it targets very old decorator and metadata specifications, not the current version. I don't super know how accurate or complete it's model really is, how closely it writes down what typescript knows versus how much it defines its own model to export to. https://www.npmjs.com/package/reflect-metadata

We are maybe at the cusp of some unifying, some coming together, albeit not via typescript at this time, still as a separate layer. The Standard Schema project has support from I dare say most of the top validation libraries. But extending this to API definitions, ORM tools is in extremely early stages. https://github.com/standard-schema/standard-schema?tab=readm...

causal3• 1 year ago

Yes. Although I am more familiar with TypeScript and prefer it for a few reasons, I've been inspired seeing how Pydantic has been used in the field for both design-time and runtime validation.

lenkite• 1 year ago

Hard agree - TSC already has this information, why should so many libraries in the world duplicate it ? A single source of truth works better than everyone repeating the same in different, finicky runtime validation libraries.

sureIy• 1 year ago

I don't want TypeScript to be a runtime checker

I do.

If the types are internally consistent, then they don't need runtime validation, it'd just be extra code to download and run.

If you accept external data, you may already opt in using something like Zod. I don't see why conflate the two things just because their quacks echo the same.

90s_dev5• 1 year ago

All APIs are experimental and in a state of constant flux and evolution, including the web.

The only time status quo is fine is when it's for clients and employers who just want stuff done and don't care how.

In that context, all these unfortunate layers of complexity at least mean more billable hours.

owebmaster2• 1 year ago

All APIs are experimental and in a state of constant flux and evolution, including the web.

for better or worse the web only adds, don't change or remove APIs

blackoil2• 1 year ago

We use drizzle + Hono + client with zod. So the entire chain from ui to client to server to DB follow same validation and shape.

causal3• 1 year ago

I've heard of this stack, glad to hear it works, still hate that you need three separate external dependencies to do it

dalmo3• 1 year ago

How do you use the same validation for drizzle and zod? I have been writing duplicate schemas and hoping they match...

tshaddox• 1 year ago

It's so frustrating how full-stack development requires so many ways of describing the same shapes: JS input validation, Swagger for API definition, server-side input validation, ORM for schema conformance, and TypeScript often requiring separate definitions on server and client side. It's so tedious.

Many of the same people who complain about how complicated modern web dev is would also shudder at the suggestion to just replace all those things with TypeScript (and Zod, if that's your TS schema definition and validation library of choice).

chamomeal• 1 year ago

You shouldn’t typically need to redefine your data all over the places. There are zillions of converters for everything to zod, and zod to everything.

If you already have JSON schema/swagger schemas, generate zod from em. If you use a typescript orm, I bet you $10 there’s a zod generator for it.

Honestly zod has gotten so popular that for me, it’s the unifying schema definition that everything else in the stack can rely on. And since directly defines the types, devs actually keep it up to date (swagger docs at my company are ALWAYS lagging behind changes)

devin• 1 year ago

Seeing an announcement for a new version of some typescript library I've never heard of shoot to the top of the front page makes me feel extremely out of touch with mainstream dev.

esafak2• 1 year ago

What do you use for validation in TS? Anyone who deals with data should be validating/parsing it.

90s_dev5• 1 year ago

  const cached = new Map<string, Function>()

  export function as<T>(o: any, path: string, as: (o: any) => T | undefined) {
    try {
      let fn = cached.get(path)
      if (!fn) cached.set(path, fn = new Function('o', `return o.${path}`))
      const v = fn(o)
      return as(v) as T | undefined
    }
    catch (e) {
      return undefined
    }
  }

  as.number = (o: any) => (typeof o === 'number' ? o : undefined)
  as.string = (o: any) => (typeof o === 'string' ? o : undefined)
  as.boolean = (o: any) => (typeof o === 'boolean' ? o : undefined)
  as.numbers = (len = 0) => (o: any) => (o instanceof Array && o.length >= len && o.every(c => typeof c === 'number') ? o : undefined)
  as.strings = (len = 0) => (o: any) => (o instanceof Array && o.length >= len && o.every(c => typeof c === 'string') ? o : undefined)


  const size = as(usrConfig, 'sys.size', as.numbers(2))
  const fontpath = as(usrConfig, 'sys.font', as.string)

strogonoff• 1 year ago

Effect Schema (the next incarnation of fp-ts/io-ts) is my go-to for parsing external data in TypeScript, and I heavily recommend it.

satvikpendem6• 1 year ago

If you don't do web dev (or at least, the latest web dev) then it's understandable, I often see lots of posts in the top spots whose languages I have no idea about either.

brulard• 1 year ago

Zod is quite common, if not the most used validation library for JS/TS projects.

varbhat• 1 year ago

I am not an expert here but I had a thought that JSON-Schema might be a good choice because since it's schema based, i can implement the validators in Non-Typescript languages too.

https://ajv.js.org is one such JSON Schema library. How does zod compare to this?

rafram7• 1 year ago

Well, Zod isn't just for validating JSON. It supports validating objects that can't be represented in JSON without transformation (dates, class instances, and so on). You can also use it as your JSON transformer if you want to - you can write your schema so it accepts strings, validates them as ISO date strings, and outputs Date objects, for instance.

nozzlegear• 1 year ago

Well, Zod isn't just for validating JSON. It supports validating objects that can't be represented in JSON without transformation (dates, class instances, and so on).

Thanks, this was the missing piece for me. I'd been thinking about using Zod for an old client's Node codebase, but I'd only need it to validate the shape of json objects received by http endpoints. Zod was looking like overkill, but I know it's popular so I wasn't sure what I was missing.

0x62• 1 year ago

Zod 4 supports converting a Zod schema to JSON-Schema (natively, this has always been possible with 3rd-party libs).

One key difference is preprocessing/refine. With Zod, you can provide a callback before running validation, which is super useful and can't be represented in JSON. This comes in handy more often than you'd think - e.g converting MM/DD/YYYY to DD/MM/YYYY before validating as date.

silverwind2• 1 year ago

Does it support the other way around too? I'd live to ditch AJV.

sroussey4• 1 year ago

It is a good choice and in that case TypeBox is a decent way to generate it.

And you can use AVJ or its own schema validation system (which is much faster, but only sync where avj has asynchronous validation options in case you want to check against a database or something).

silverwind2• 1 year ago

JSON Schema is the way for cross-language validation. Wrap it in OpenAPI and you also get nice API docs for free.

ar-nelson• 1 year ago

Obligatory shameless plug whenever Zod is posted: if you want similar, but much more minimal schema validation at runtime, with a JSON representation, try Spartan Schema: https://github.com/ar-nelson/spartan-schema

jilles• 1 year ago

When I see a repository with many files "updated 4 years ago" I'm usually inclined to think it's abandoned.

ashwinsundar2• 1 year ago

Or the code is done and doesn’t need to be iterated on continuously?

johnfn6• 1 year ago

I'm curious if anyone here can answer a question I've wondered about for a long time. I've heard Zod might be in the right ballpark, but from reading the documentation, I'm not sure how I would go about it.

Say I have a type returned by the server that might have more sophisticated types than the server API can represent. For instance, api/:postid/author returns a User, but it could either be a normal User or an anonymous User, in which case fields like `username`, `location`, etc come back null. So in this case I might want to use a discriminated union to represent my User object. And other objects coming back from other endpoints might also need some type alterations done to them as well. For instance, a User might sometimes have Post[] on them, and if the Post is from a moderator, it might have special attributes, etc - another discriminated union.

In the past, I've written functions like normalizeUser() and normalizePost() to solve this, but this quickly becomes really messy. Since different endpoints return different subsets of the User/Post model, I would end up writing like 5 different versions of normalizePost for each endpoint, which seems like a mess.

How do people solve this problem?

peab• 1 year ago

Either Unions, or optional fields

adpirz• 1 year ago

It's hard to unpack without knowing more about the use case, but adding discriminant properties (e.g. "user_type") to all the types in the union can make it easier to handle the general and specific case.

E.g.

if (user.user_type === 'authenticated') {

  // do something with user.name because the type system knows we have that now

}

probabletrain3• 1 year ago

In an ideal world you'd have one source of truth for what the shape of a User could be (which may well be a discriminated union of User and AnonymousUser or similar).

Without fullstack TS this could look something like: (for a Python backend) Pydantic models+union for the various shapes of `User`, and then OpenAPI/GraphQL schema generation+codegen for the TS client.

johnfn6• 1 year ago

The problem with this is that your One True User shape tends to have a bunch of optional properties on it. e.g., in the user profile you fetch Post[], but in the user directory you don't, and so on with other joined properties. If every endpoint returns the One True User, then you end up needing to write conditional logic to guard against (say) `.posts` when you fetch users in the profile, even though you know that `.posts` exists.

stephen_hn2• 1 year ago

I think you would use discrimated unions.

const MyResult = z.discriminatedUnion("status", [ z.object({ status: z.literal("success"), data: z.string() }), z.object({ status: z.literal("failed"), error: z.string() }), ]);

You can define passthrough behavior if there are a bunch of special attributes for a moderator but you don't want to list/check them all.

With different methods that have different schema- If they share part of the same schema with alterations, you can define an object for the shared part and create objects that contain the shared object schema with additional fields.

If you have a lot of different possibilities, it will be messy, but it sounds like it already is for you, so validators could still at least validate the messines.

johnfn6• 1 year ago

Hmm, this is pretty interesting, but I still have issues. While Zod seems to be able to transform my server-returned object into a discriminated union, it doesn't seem to be able to tell which of the unions it is. For instance, say that api/user/mod-panel (e.g.) returns only moderator user objects. My API endpoint already knows that it will have moderator-related fields, but Zod forgets that when I parse it.

gavinray• 1 year ago

This doesn't help in your case, but this is what GraphQL was invented for.

TS libraries for GQL queries can dynamically infer the response shape from the shape of the selected fields.

Such that the query

    query {
        user {
            username
            posts {
                text
            }
        }
    }

Would be

    type Response = {
        user: {
            username: string
            posts: Array<{
                text: string
            }>
        }
    }

And a query with just { user { username } } would have the posts property omitted entirely.

mmcnl2• 1 year ago

Your server should export the types. Don't write types by hand in your client, that makes no sense. The server knows what data it is sending and should provide that information as metadata to the client. Practically this can mean that a Python backend uses Pydantic schemas which can be used to automatically generate an OpenAPI specification that your client can use to generate types.

johnfn6• 1 year ago

This answer is how I wish the world would work, but I am stuck with a server with a poor type system that can't refine the types nearly as accurately as TS can. :(

satvikpendem6• 1 year ago

Zod 4 looks good but even with their latest improvements, ArkType is still an order of magnitude faster. Sometimes for the sake of backward and syntax compatibility, it is difficult to make something much faster than a fully greenfield newer library. We recently did an analysis of all these types of tools for our project and decided to go with ArkType for partially this reason, the other was TypeScript ergonomics felt nicer.

atonse• 1 year ago

I was looking at all their speed metrics, and can you explain to me where the speed makes a difference?

We only use zod to validate forms, so I keep thinking "how does this matter?" Are people maybe using it to validate high throughput API input messages or something like that, where performance may matter more?

satvikpendem6• 1 year ago

Yes, we use it to validate the API responses from the backend (as well as type validation on the backend itself from any frontend POST requests), and especially on the client side, speed and bundle size is very important.

andrewflnr• 1 year ago

Interesting, that one didn't even appear in my research. I was specifically looking for typescript ergonomics too. Probably still not going to switch away from Zod, though.

sroussey4• 1 year ago

Why ArkType over TypeBox?

dangoodmanUT• 1 year ago

ArkType is a nightmare to use, zod is nice to use

lhnz• 1 year ago

Out of interest, why is it a nightmare to use?

I've always been worried about how overly clever the approach is, does it have problems?

indigovole• 1 year ago

With yet another exciting new release of something reaching the top of HN, I would just like to urge devs to put a description of the project they're actually releasing and a link to the page describing the project in the release announcement.

These announcements could be a valuable touchpoint for you to reach a whole new audience, but I can't remember a single one that starts with something like "exciting new release of NAME, the X that does Y for users of Z. Check out the project home page at https:// for more."

Quite often, the release announcement is a dead end that can't even take me to the project! In this case, the only link is a tiny octocat in the lower left-hand corner, AFAICS.

rossant• 1 year ago

I came here to say this. I clicked on the link and had no idea what this was about.

pikelet• 1 year ago

The next step is to click the project logo in the top left.

8s2ngy• 1 year ago

Congratulations to the Zod team on the new release. At the risk of sounding overtly negative, I can't help but shudder when I think about the number of breaking changes outlined in the migration guide. For projects that rely heavily on Zod, it feels like a daunting task ahead—one that will demand a lot of developer attention and time to navigate. Having maintained a few frontend projects that are 4-5 years old at work, I really empathize with them.

In my experience, large React projects often depend on a multitude of libraries, and when each one rolls out substantial changes—sometimes with barely any documentation—it can quickly become overwhelming. This is honestly one of my least favorite aspects of working with JavaScript. It just feels like a constant uphill battle to keep everything in sync and functioning smoothly.

koakuma-chan4• 1 year ago

For projects that rely heavily on Zod, it feels like a daunting task ahead—one that will demand a lot of developer attention and time to navigate.

Or just use an LLM.

camgunz• 1 year ago

And then pick through all the LLM's mistakes.

owebmaster2• 1 year ago

better yet: use an LLM to generate a subset of Zod that is fit for the project

cscheid• 1 year ago

This is the kind of task that LLMs are precisely terrible at; there isn't an abundance of Zod 4 examples, and the LLM will sure as shit will give you _something_ you are now by definition ill-equipped to assess.

I'm confident about this assessment because I maintain a large-ish piece of software and perenially have to decipher user reports of hallucinated LLM syntax for new features.

sensanaty• 1 year ago

LLMs can't even not write React when you explicitly tell them not to write React in your Vue codebase. I still, with Gemini 2.5 Pro and Claude 3.7 both, get this annoying ass interaction almost daily, despite a universe of context there that should make it obvious I don't want a React component in my Vue codebase.

ruined2• 1 year ago

that's why they're taking the dual-availability approach, with a separate 'mini' edition. it's easy to perform a progressive migration without messing with the package manager.

consumers uninterested in the 'mini' edition don't have to bother with that part.

but, the benefits of the 'mini' edition are so drastic for tree-shaking that it was driving development of alternatives - zod had to either do it (and benefit), or deprecate.

JoRyGu• 1 year ago

Was just looking at their release strategy. This is being handled by people that have experienced the hell that is dependency management in the JS ecosystem. Kudos to them.

nicksergeant• 1 year ago

I am 100% in agreement here. I operate a couple large Next.js apps and in the last year we've had to deal with Next.js 14 -> 15 which introduced a ton of breaking cache changes, Next.js pages -> app router, React 18 -> 19, Eslint 8 -> 9, and Tailwind 3 -> 4.

It's honestly been a nightmare, and I wish I had just built in Django instead. The Tailwind 3 -> 4 migration was probably among the most painful, which I was not expecting.

Daishiman• 1 year ago

I will continue to beat the drum that using easy-to-migrate frameworks like Django that don't go around introducing major breaking changes is one of the bigger web development superpowers.

winstonp• 1 year ago

i am just simply not touching tailwind v3 -> 4. v4 is for new projects only.

egorfine2• 1 year ago

OPINION WARNING

Eslint 8 → 9 was not just incredibly painful with a shock wave that still propagates through the industry, but it was also totally useless

rafram7• 1 year ago

To simplify the migration process both for users and Zod's ecosystem of associated libraries, Zod 4 is being published alongside Zod 3 as part of the zod@3.25 release. [...] import Zod 4 from the "/v4" subpath

npm is an absolute disaster of a dependency management system. Peer dependencies are so broken that they had to make v4 pretend it's v3.

fhd2• 1 year ago

I feel like both Node.js and NPM were initially vibe coded before LLMs existed. Just a quick hack, kind of, that got hugely popular somewhat by accident.

Edit: Thinking about it, that's the origin story of JavaScript as well, so rather fitting.

derN3rd• 1 year ago

We've been using zod 4 beta already with great improvements but due to our huge codebase not being able to handle the required moduleResolution settings, we cannot upgrade...

They could at least also publish it as a major version without the legacy layer

EDIT: I've just seen the reason described here: https://github.com/colinhacks/zod/issues/4371 TLDR: He doesn't want to trigger a "version bump avalanche" across the ecosystem. (Which I believe, wouldn't happen as they could still backport fixes and support the v3 for a time, as they do it right now)

sroussey4• 1 year ago

Would not be hard for you or anyone else to do that as well.

827a• 1 year ago

Yeah this seems like very poor reasoning tbh.

gejose2• 1 year ago

Peer dependencies are so broken that they had to make v4 pretend it's v3

I'm not sure this is the right conclusion here. I think zod v4 is being included within v3 so consumers can migrate over incrementally. I.e refactor all usages, one by one to `import ... from 'zod/v4'`, and once that's done, upgrade to v4 entirely.

skydhash7• 1 year ago

How do you even migrate incrementally? why keep both old and new code together like a Frankeinstein project? Especially on a codebase you own. It's a library, not a platform.

paulddraper4• 1 year ago

But that was always a choice.

You can import aliases.

If you have a code that needs to use two versions of axios, or zod, or whatever...

  "zod4": "npm:zod@4.0.0"

herrkanin• 1 year ago

I might be blindsided by using npm exclusively for years by this point, but what would be a better way to support iteratively migrating from v3 to v4 without having to do it all in one large batch?

strken• 1 year ago

Using npm's built in support for package aliases to simultaneously install zod@3 as zod and zod@4 as zod-next?

Edit: reading the rationale, it's about peer dependencies rather than direct dependencies. I am still a little confused.

diggan4• 1 year ago

One possible solution: Publish a new package with a new name. I've personally been thinking of doing that for some libraries, where I'd commit to never change the public API, and if I figure out a nicer way of doing something, I'd create a new library for it instead, and people who wanna use/maintain the old API can continue to do so.

swyx• 1 year ago

apart from the other things people raised, as leerob also points out it hints to an llm that the api may be different https://x.com/leerob/status/1924504939818008791

Trufa• 1 year ago

I feel like people will upvote anything negative. This is not much a limitation of npm, there's nothing intrinsically wrong with npm that lead to this decision, this is more a pragmatic way of allowing progressive change of a library that introduced a lot of braking changes.

danenania• 1 year ago

Right, you have the same issues to consider when shipping a breaking major version upgrade to a new library in any language/ecosystem.

That said, you do see a cultural difference in node-land vs. many other ecosystems where library maintainers are much quicker to go for the new major version vs. iterating on the existing version and maintaining backward compatibility. I think that's what people are mostly really referring to when they complain about node/npm.

Webpack is a good example—what's it on now, version 5? If it was a Go project, it would probably still be on v1 or maybe v2. While the API arguably does get 'better' with each major version, in practice a lot of the changes are cosmetic and unnecessary, and that's what frustrates people I think. Very few people really care how nice the API of their bundler is. They just want it to work, and to receive security updates etc. Any time you spend on upgrading and converting APIs is basically a waste—it's not adding value for users of your software.

colinmcd12• 1 year ago

Author here. I wrote a fairly detailed writeup here[0] for those who are interested in the reasons for this approach.

Ultimately you're right that npm doesn't work well to manage the situation Zod finds itself in. But Zod is subject to a bunch of constraints that virtually no other libraries are subject to. There are dozens or hundreds of libraries that directly import interfaces/classes from "zod" and use them in their own public-facing API.

Since these libraries are directly coupled to Zod, they would need to publish a new major version whenever Zod does. That's ultimately reasonable in isolation, but in Zod's case it would trigger a "version avalanche" would just be painful for everyone involved. Selfishly, I suspect it would result in a huge swath of the ecosystem pinning on v3 forever.

The approach I ended up using is analogous to what Golang does. In essence a given package never publishes new breaking versions: they just add a new subpath when a new breaking release is made. In the TypeScript ecosystem, this means libraries can configure a single peer dependency on zod@^3.25.0 and support both versions simultaneously by importing what they need from "zod/v3" and "zod/v4". It provides a nice opt-in incremental upgrade path for end-users of Zod too.

[0] https://github.com/colinhacks/zod/issues/4371

codeflo• 1 year ago

Is there any advantage to this approach over publishing a separate "zod4" package? That would be just as opt-in and incremental, not bloat download sizes forever, and make it easier to not accidentally import the wrong thing.

Lvl999Noob• 1 year ago

Are there any plans of doing an actual v4 release and making zod/v4 the default there? Perhaps make simulatenous releases of zod v3 containing a /v4 path and a zod v4 containing a /v3 path? Then converge on just zod v4 with no /v3 from 4.1 onwards.

bilalq2• 1 year ago

This isn't an npm exclusive issue. A dependency having some other transitive deps also depend on an older version is a problem that happens in literally every other ecosystem. If anything, npm gives you more escape hatches by actually allowing you to run multiple versions concurrently if you need to or selectively overriding parts of your transitive dependency graph.

What package management system has a solution to this? Even so called "stable" platforms like Maven deal with this nonsense by publishing new versions under a new namespace (like Apache Commons did from v2 to v3).

aiiizzz• 1 year ago

Horrible, not going to use it now. My ide will import whichever interchangeably, and it'll be a mess

revskill• 1 year ago

Mind blowing.

mac9• 1 year ago

The little animation that highlights what section you are reading currently is beautiful

johnisgood• 1 year ago

What animation? I actually do not see it.

yhprum2• 1 year ago

This is a feature of Fumadocs that the site uses!

[1] https://fumadocs.dev/

jasonthorsness• 1 year ago

But how is the below possible - doesn't it need to include most of the TypeScript compiler? Does it compile the type definitions to some kind of validation structure (like how a compiled regex works) and then use just that?

    - Zero external dependencies
    - Works in Node.js and all modern browsers
    - Tiny: 2kb core bundle (gzipped)

_tqr32• 1 year ago

The TypeScript compiler is only needed during development, the compiled JavaScript code contains no TypeScript-specific logic. Zod’s validation is entirely JavaScript-based, relying on simple checks (e.g., typeof, regexes, comparisons).

Also, the 2kb bundle just for the zod/v4-mini package, the full zod/v4 package is quite large.

hdjrudni5• 1 year ago

5.36kB is not "quite large". It's pretty reasonable, esp. if you're only running it on the server.

Aeyxen• 1 year ago

I think Zod uses JIT compilation via `new Function`, rather than including the entire TypeScript compiler. This method allows for concise validation logic, executing only what’s necessary at runtime.

_tqr32• 1 year ago

Still no hapi support?

h1fra• 1 year ago

Impressive results, we need some deep dive on the performance improvements!

labadal• 1 year ago

I just got started working Zod into a new project. This could not have happened at a better time. I would have needed to change so much ot migrate to v4 based on what I'm seeing.

nikolay3• 1 year ago

Zod and the way to define schemas with it has been one of the reasons I didn't want to touch TypeScript as it's becoming more prevalent!

rlt• 1 year ago

Can you elaborate?

nikolay3• 1 year ago

It defines yet another way to declare schemas in a way that's not reusable. ArkType mostly uses TypeScript's, so, it feels more natural.

aduffy• 1 year ago

Seeing v4-mini reminds me of an LLM. I can only assume the next major release will be zod-3.9-medium-high

TheFlashBold• 1 year ago

Or maybe 3.8-q4_K_M for a quantized version

dankobgd• 1 year ago

i would use it if there was just 1 normal zod lib not like how it is now: zod and zod mini and who knows what. I rather use arktype or valibot if i am in browser only.

sroussey4• 1 year ago

Hopefully someone can compare Zod4 to TypeBox.

Last I looked, the nice thing about TypeBox was that is _was_ JsonSchema just typed which was nice for interoperability.

hdjrudni5• 1 year ago

I was using TypeBox but switched to Zod. I like TypeBox but I found I don't actually need my schema to be a JSON Schema and the one time I did want a JSON Schema it was hard to get actual JSON back out of TypeBox.

The thing I like about Zod is the "refinements" and "transforms" and the new "overwrite" in Zod 4 was just what I wanted.

TypeBox does have a transform (https://github.com/sinclairzx81/typebox?tab=readme-ov-file#t...) but you can't transform and then apply more validations and then maybe another transform, etc. For example, I like to trim strings before checking length.

baalimago• 1 year ago

I read this, and I find myself not wanting to migrate since I know that LLMs can't generate Zod 4 syntax yet.

punkpeye• 1 year ago

Shocked by the negativity around this.

I tested early versions of zod v4 and liked the new API, but was very concerned about what will be the migration path. I was even going to suggest to publish under a new package name.

But the author's approach is ingenious. It allows for someone like me to start adopting v4 immediately without waiting for every dependency to update.

Well done!

hombre_fatal3• 1 year ago

Yeah, it's a nice route. There's no way I could afford to do an all or nothing migration for something like validation.

colinmcd12• 1 year ago

Author here, AMA!

Regarding the versioning: I wrote a fairly detailed writeup here[0] for those who are interested in the reasons for this approach.

Ultimately npm is not designed to handle the situation Zod finds itself in. Zod is subject to a bunch of constraints that virtually no other libraries are subject to. Namely, the are dozens or hundreds of libraries that directly import interfaces/classes from Zod and use them in their own public-facing API.

[0] https://github.com/colinhacks/zod/issues/4371

miohtama• 1 year ago

The path of least short term pain is often the best path.

Everyone old in Python ecosystem remembers the Python 2/3 migration madness.

elbajo2• 1 year ago

First thank you for the hard work, many of my local hacks will go away with the new features!

As a convenience and mostly avoid typos in form names I use my own version of https://github.com/raflymln/zod-key-parser. I've been surprised something like this hasn't been implemented directly in the library.

Curious if you think this is out of scope for Zod or just something you haven't gotten around to implement?

(Here are discussions around it: https://github.com/colinhacks/zod/discussions/2134)

colinmcd12• 1 year ago

Some kind of affordance for FormData/URLSearchParams-style structures is definitely in scope. It was a late cut. Ultimately HTML inputs/forms are an implicit type system unto itself—certainly HTML has a very different notion of "optional" than TypeScript. So my approach to this would likely involve another sub-library ("zod/v4-form").

I like your library! Put in a PR to add it to the ecosystem page :)

transitivebs• 1 year ago

this blog post was a delight to read. great work, colin && thanks!!

jjice• 1 year ago

Big thank you for the work on Zod - it's a fantastic library and it's been incredibly valuable for data validation/parsing in Node environments for me over the last two years.

paulddraper4• 1 year ago

Am I to understand the earliest version of Zod 4 is 3.25.0?

Does this not sound insane?

---

I've been using the alpha versions of Zod for months, I just want to edit package.json and upgrade. But now I need to shotgun across git history instead.

Colin, I appreciate your project immensely. As a point of feedback, you made this ^ much harder than you had to. (Perhaps publish a 4.x along with the 3.x stuff?)

colinmcd12• 1 year ago

I understand this as a knee-jerk reaction. I didn't do this lightly.

Perhaps publish a 4.x along with the 3.x stuff

You have some misconceptions about how npm works. Unfortunately it's less reasonable than you think. There's a single `latest` tag, and there's only one "latest" version at a time. It's expected that successive versions here will follow semver. Once I publish zod@4 I can no longer publish additional zod@3.x.x versions. The workaround here is to publish v3 versions to a separate dist tag (zod@three) but anyone consuming that dist-tag (e.g. "zod": "three" in their package.json) loses the ability to specify a semver range.

I recommend reading the writeup[0]. I don't think you're appreciating the magnitude of the disruption a simple major version bump would have caused, or the reasons why this approach is necessary to unlock continuity for Zod's ecosystem libraries. They're quite subtle.

[0] https://github.com/colinhacks/zod/issues/4371

Shacklz• 1 year ago

Many thanks for your work! Definitely looking forward to the upgrade - especially the performance-boosts with regards to tsc will be very welcome in our relatively large code-base, and the changes to discriminated unions will probably help us big time in a very specific scenario where so far they fell short.

That being said, I'm fully understanding of the reasons for the somewhat odd versioning given your special situation, but still, I'd wish there would be a 4.0.0-package for folks like us who simply don't need to worry or bother about zod-version-clashes in transitive dependencies bacause those don't exist (or at least I think so; npm ls zod only returns our own dependency of zod). If I understood correctly, we'll need to adapt the import to "zod/v4", which will be an incredibly noisy change and will probably cause quite a few headaches when IDEs auto-import from 'zod' and such, which we then need to catch with linting-rules.

But that's probably a small gripe for a what sounds overall like a very promising upgrade - many thanks for your work once again!

e1g• 1 year ago

+1. We use Zod only for internal validation and would love a standalone release for v4 (or another package like zod4 or @zod/zod). I'm sure many greenfield projects would prefer that, too.

kaoD• 1 year ago

Sorry if it's mentioned in the article but I'm on mobile.

Is fixing .optional() in TS[0] part of the 9/10 top-issues fixed? This has been my biggest pain point with Zod... but still Zod is so good I still choose to just deal with it :) Thanks for an amazing part of the ecosystem.

[0] https://github.com/colinhacks/zod/issues/635

colinmcd12• 1 year ago

Yes! Zod now differentiates between `z.string().optional()` and `z.union([z.string(), z.undefined()])` (as in TypeScript itself). Details: https://x.com/colinhacks/status/1919291504587137496

waynenilsen• 1 year ago

I had so much difficulty with recursive types mixed with discriminated union (think xml in json for example) hopefully it'll be better now

jmull• 1 year ago

A perfectly valid evolution strategy is to add new APIs while keeping the old (which you may or may not choose to deprecate). That is, add the new APIs to 3.x (but probably call them something other than v4.

Now, that's effectively what 3.25 is. But there are some problematic extras... the semantics of the semantic versioning is muddled. That means confusion, which means some people will spin their wheels after not initially grokking the situation correctly. Also, there seems to be an implication there could be a strong deprecation of v3 APIs coming. That is, you have to wonder how long the window for incremental migration will remain open.

To be frank, I wouldn't touch zod on any project I hoped will be around for a while. Predicting the future is an uncertain business, but we can look at the past. In a few years, I think it's reasonable to guess zod v4 will be getting the same treatment zod v3 is getting now.

Not that I think you ought to do anything different. I probably wouldn't want to maintain some old API I came up with years ago indefinitely either (not without a decent support contract, that is).

dclowd9901• 1 year ago

Just would like to say I really appreciate your consideration of releasing breaking changes. I know not all libraries can employ your methodology but I wish many more of them would, as a frontend platform engineer.

ssousa666• 1 year ago

I saw your tweet about cracking type inference for recursive schemas. It seems like the solution to this problem is pretty straightforward and simple. Was the solution really as simple as using a getter, or is there some additional complexity that I am failing to grok?

mary-ext2• 1 year ago

the solution is to avoid TypeScript eagerly validating the object's value type. eager validation breaks cyclic references.

bilalq2• 1 year ago

While publishing both versions together helps with compatibility issues, the library now becomes even heavier in environments where tree-shaking is not enabled. In React Native, for example, you have to jump through a lot of hoops to get tree-shaking to work (including getting off the default metro bundler, which can often be a non-starter).

nisten• 1 year ago

TLDR: What do I put on my package json to get latest latest latest? "^4" ?

egorfine2• 1 year ago

A heartfelt thank you for not breaking the world like many other libraries do!

Breaking changed in fundamental but auxiliary libraries are so painful in the world of frontend development that it's not even funny anymore.

robertlagrant• 1 year ago

The approach I ended up using is analogous to what Golang does. In essence a given package never publishes new breaking versions: they just add a new subpath when a new breaking release is made. In the TypeScript ecosystem, this means libraries can configure a single peer dependency on zod@^3.25.0 and support both versions simultaneously by importing what they need from "zod/v3" and "zod/v4". It provides a nice opt-in incremental upgrade path for end-users of Zod too.

This is extremely sensible. And it means you can provide security updates for older versions, just all in the same codebase's releases.

IshKebab• 1 year ago

Uhm... Wouldn't it make more sense just to publish a `zod4` package?

andrewingram2• 1 year ago

Is there any interop between v3 and v4 schemas themselves? We have an enormous graph of hundreds (if not thousands) of Zod schemas, many inheriting or extending others. It's difficult to envisage us being able to do an incremental upgrade unless both versions can interact neatly.

But I am looking forward to seeing if this fixes our issue where in the worst case a 500 line schema file turns into 800,000 lines of generated types.

paulddraper4• 1 year ago

No. You will have errors if you try this.

colinmcd12• 1 year ago

No, that kind of interop, especially static interop (assignability), would've been totally unworkable. Despite the length of the changelog, there are very few breaking changes to the user-facing API surface. It's mostly internal/structural changes and deprecations (most of which can be fixed with a find&replace).

Report back about that .d.ts issue. It should be far better. That kind of type explosion usually happens when TypeScript needs to infer function/method return types. Zod 4 uses isolatedDeclarations so this kind of thing shouldn't happen.

__natty__• 1 year ago

Great job! I love support of json conversion you built-in. However, I find top level string bad choice as it was more verbose before, to have them under string() namespace.

bobbyraduloff• 1 year ago

great job on zod. it’s an incredible library. also really excited about the locale errors. finally I can push the zod errors directly to the frontend. would love to take responsibility for the BG locale.

msukkarieh• 1 year ago

We've been using zod for our open source tool and are loving it! Will definitely try to migrate over to v4 soon, thanks for the project!

CGamesPlay2• 1 year ago

One decision that seems dubious to me is the zod/v4-mini import. I suspect that this will actually increase bundle sizes across the ecosystem. The docs specifically say "zod/v4 is still recommended for the majority of use cases", so application developers will use it. Library authors will think to themselves, "but I want to enable my library to be used by those with uncommonly strict bundle size requirements", and so will use that. The net result is that both zod/v4 and zod/v4-mini will be included in the application bundle.

I guess this could be mostly avoided if zod/v4 actually a wrapper around zod/v4-mini. Is that the case?

LinusU• 1 year ago

Library authors should only import from `core`, which is used by both v4 and v4-mini.

ref: https://zod.dev/library-authors?id=how-to-support-zod-and-zo...

CGamesPlay2• 1 year ago

Glad to see that this was also considered. I’m curious why they went with a third API, though.

s-mon• 1 year ago

Who else here is going to the Zod meetup tonight?

LMMojo• 1 year ago

All that text. So, so much text and not one single (maybe there was, but I fell asleep), not one single "Let me tell you what Zod is" paragraph. Lots of "Zod 4 is better than Zod 3" and "Here's what we pulled back, out, in, from Zod 2".

ruined2• 1 year ago

there's an 'Intro' link under the 'Documentation' header on the left side

aobdev• 1 year ago

"TypeScript-first schema validation with static type inference"

MobiusHorizons• 1 year ago

The website fails to load with a javascript error in Safari.

`SyntaxError: Invalid regular expression: invalid group specifier name`

lioeters• 1 year ago

Looks like this feature is not yet supported in Safari.

Lookbehind in JS regular expressions - https://caniuse.com/js-regexp-lookbehind

mrwww• 1 year ago

Defining schemas in zod and using z.infer<> to get my types is just lovely, thank you!

_1tem5• 1 year ago

As a full stack dev that runs their own SaaS with thousands of users in production, sometimes I get a very pleasant reminder of how many problems I am blissfully unaware of just because I decided not to build an SPA or use JS frontend frameworks. It never occurred to me that I need such a thing as Zod/ArkType - first time hearing of it, and I have no use for it.

It boggles my mind how much effort and complexity and tooling goes into building an SPA. Entire classes of problems simply don't exist if you choose not to build an SPA. Meanwhile, I use the browser as designed: with full page reloads, backend development only, and occasional reactivity using a backend-only framework like Laravel Livewire. Everything is so simple: from access control to validation to state management. And yes, my app is fast, reactive, modern, SEO friendly, and serves thousands of users in production.

giorgioz• 1 year ago

What's your SaaS? I would like to check out your use case that allowed you to build a great product without SPA.

_1tem5• 1 year ago

Basecamp and Hey are $100m+ software companies that use Ruby on Rails without an SPA.

SebastianKra• 1 year ago

Well how do you do server-side validation?

Because now I have the impression that you're just defensively writing a bunch of if-statements. Or worse, you rely on html client-side form validation only.

_1tem5• 1 year ago

Any good backend framework (Laravel, Django, Livewire, Rails) does server side validation out of the box without me as a developer needing to manually pass around JSON objects and continually checking their shape. The boundary between user input and backend code is well defined and border checks are easy.

zachrip• 1 year ago

Zod has nothing to do with SPAs and isn't just a frontend tool. Do you not validate inputs in your code?

_1tem5• 1 year ago

Laravel/Livewire handles it. I have no idea why something like Zod should be the concern for a high level application dev. For a framework dev, sure

dimitrisnl• 1 year ago

Entire classes of problems simply don't exist, if you also close your eyes.

andrewingram2• 1 year ago

We have an enormous number of backend-only Zod schemas that are used as part of our data pipelines; so it doesn't seem like something that only exists to solve SPA problems.

_1tem5• 1 year ago

If it’s backend only why do JSON shapes need to be continually checked? Can’t you serialize higher level classes or use the database?

tauchunfall4• 1 year ago

I worked on two large projects that use Zod and Protobuf to ensure schema evolution goes well with 5+ teams. Even if you have a lightweight frontend, it makes sense to use something that validates the schemas in the backend.

In the end you'll have a schema somewhere. Maybe defined in the database system, maybe defined in the models of an object-relational mapper. Since defining the schemas in database system or using object-relational mappers can cause very difficult problems in large projects we do not use it and use Zod and Protobuf instead. While I think you could even replace Protobuf completely with Zod or something similar.

There is the British Post Office scandal [1] around an IT system called Horizon (the legacy system in this case). After reading about the details I'm pretty sure something like Zod (I mean any schema validation) would have contributed to prevent this scandal. They even mentioned the lack of using a schema in the technical appendices in the court documents of the group legal action [2].

[1] https://en.wikipedia.org/wiki/British_Post_Office_scandal

[2] https://en.wikipedia.org/wiki/Bates_%26_Others_v_Post_Office...

slackfan• 1 year ago

hmmm. Zod on Zed?

pipallweek• 1 year ago

The versioning approach here is a really interesting compromise — especially in the npm ecosystem where breaking changes ripple out so painfully through transitive dependencies. From a developer ergonomics standpoint though, the shift to zod/v4 imports will definitely create some friction. For teams with IDEs auto-importing from 'zod' and linters enforcing import styles, it might introduce subtle DX issues until workflows are adjusted.

That said, the strategy does seem to prioritize ecosystem stability over short-term convenience, which is fair. Would love to see better tooling (maybe even IDE plugins or codemods) to help projects transition cleanly. Really appreciate the thoughtful design behind all of this.

assimpleaspossi2• 1 year ago

Wow! This is great!!!

What is it?

bdangubic• 1 year ago

Zod, version 4

assimpleaspossi2• 1 year ago

Which tells me less than what the link to the web site tells me.

enbugger• 1 year ago

The library pretty much remained unusable garbage without documentation comments. I have no idea why it became so popular and why this release is named "stable" when there is even no 4.0.0.

Zod 4

Related Stories