Set up with Next.js App Router
On this page
Recommended file structure
.
├── src
│ ├── app
│ │ ├── api
│ │ │ └── trpc
│ │ │ └── [trpc]
│ │ │ └── route.ts # <-- tRPC HTTP handler
│ │ ├── layout.tsx # <-- mount TRPCReactProvider
│ │ └── page.tsx # <-- server component
│ ├── trpc
│ │ ├── init.ts # <-- tRPC server init & context
│ │ ├── routers
│ │ │ ├── _app.ts # <-- main app router
│ │ │ ├── post.ts # <-- sub routers
│ │ │ └── [..]
│ │ ├── client.tsx # <-- client hooks & provider
│ │ ├── query-client.ts # <-- shared QueryClient factory
│ │ └── server.tsx # <-- server-side caller
│ └── [..]
└── [..]Add tRPC to an existing Next.js App Router project
1. Install deps
pnpm add @trpc/server @trpc/client @trpc/tanstack-react-query @tanstack/react-query@latest zod client-only server-onlynpm install @trpc/server @trpc/client @trpc/tanstack-react-query @tanstack/react-query@latest zod client-only server-onlyyarn add @trpc/server @trpc/client @trpc/tanstack-react-query @tanstack/react-query@latest zod client-only server-onlybun add @trpc/server @trpc/client @trpc/tanstack-react-query @tanstack/react-query@latest zod client-only server-only2. Create a tRPC router
Initialize your tRPC backend in trpc/init.ts using the initTRPC function, and create your first router. We're going to make a simple "hello world" router and procedure here — for deeper information on creating your tRPC API, refer to the Quickstart guide and Backend usage docs.
import { initTRPC } from '@trpc/server';
/**
* This context creator accepts `headers` so it can be reused in both
* the RSC server caller (where you pass `next/headers`) and the
* API route handler (where you pass the request headers).
*/
export const createTRPCContext = async (opts: { headers: Headers }) => {
// const user = await auth(opts.headers);
return { userId: 'user_123' };
};
// Avoid exporting the entire t-object
// since it's not very descriptive.
// For instance, the use of a t variable
// is common in i18n libraries.
const t = initTRPC
.context<Awaited<ReturnType<typeof createTRPCContext>>>()
.create({
/**
* @see https://trpc.io/docs/server/data-transformers
*/
// transformer: superjson,
});
// Base router and procedure helpers
export const createTRPCRouter = t.router;
export const createCallerFactory = t.createCallerFactory;
export const baseProcedure = t.procedure;import { z } from 'zod';
import { baseProcedure, createTRPCRouter } from '../init';
export const appRouter = createTRPCRouter({
hello: baseProcedure
.input(
z.object({
text: z.string(),
}),
)
.query((opts) => {
return {
greeting: `hello ${opts.input.text}`,
};
}),
});
// export type definition of API
export type AppRouter = typeof appRouter;3. Create the API route handler
With the App Router, use the fetch adapter to handle tRPC requests. Create a route handler that exports both GET and POST:
import { fetchRequestHandler } from '@trpc/server/adapters/fetch';
import { createTRPCContext } from './trpc/init';
import { appRouter } from './trpc/routers/_app';
const handler = (req: Request) =>
fetchRequestHandler({
endpoint: '/api/trpc',
req,
router: appRouter,
createContext: () => createTRPCContext({ headers: req.headers }),
});
export { handler as GET, handler as POST };4. Create a Query Client factory
Create a shared file trpc/query-client.ts that exports a function that creates a QueryClient instance.
import {
defaultShouldDehydrateQuery,
QueryClient,
} from '@tanstack/react-query';
import superjson from 'superjson';
export function makeQueryClient() {
return new QueryClient({
defaultOptions: {
queries: {
staleTime: 30 * 1000,
},
dehydrate: {
// serializeData: superjson.serialize,
shouldDehydrateQuery: (query) =>
defaultShouldDehydrateQuery(query) ||
query.state.status === 'pending',
},
hydrate: {
// deserializeData: superjson.deserialize,
},
},
});
}We're setting a few default options here:
staleTime: With SSR, we usually want to set some default staleTime above 0 to avoid refetching immediately on the client.shouldDehydrateQuery: This is a function that determines whether a query should be dehydrated or not. Since the RSC transport protocol supports hydrating promises over the network, we extend thedefaultShouldDehydrateQueryfunction to also include queries that are still pending. This will allow us to start prefetching in a server component high up the tree, then consuming that promise in a client component further down.serializeDataanddeserializeData(optional): If you set up a data transformer in the previous step, set this option to make sure the data is serialized correctly when hydrating the query client over the server-client boundary.
5. Create a tRPC client for Client Components
The trpc/client.tsx is the entrypoint when consuming your tRPC API from client components. In here, import the type definition of
your tRPC router and create typesafe hooks using createTRPCContext. We'll also export our context provider from this file.
'use client';
// ^-- to make sure we can mount the Provider from a server component
import type { QueryClient } from '@tanstack/react-query';
import { QueryClientProvider } from '@tanstack/react-query';
import { createTRPCClient, httpBatchLink } from '@trpc/client';
import { createTRPCContext } from '@trpc/tanstack-react-query';
import { useState } from 'react';
import { makeQueryClient } from './query-client';
import type { AppRouter } from './routers/_app';
export const { TRPCProvider, useTRPC } = createTRPCContext<AppRouter>();
let browserQueryClient: QueryClient;
function getQueryClient() {
if (typeof window === 'undefined') {
// Server: always make a new query client
return makeQueryClient();
}
// Browser: make a new query client if we don't already have one
// This is very important, so we don't re-make a new client if React
// suspends during the initial render. This may not be needed if we
// have a suspense boundary BELOW the creation of the query client
if (!browserQueryClient) browserQueryClient = makeQueryClient();
return browserQueryClient;
}
function getUrl() {
const base = (() => {
if (typeof window !== 'undefined') return '';
if (process.env.VERCEL_URL) return `https://${process.env.VERCEL_URL}`;
return 'http://localhost:3000';
})();
return `${base}/api/trpc`;
}
export function TRPCReactProvider(
props: Readonly<{
children: React.ReactNode;
}>,
) {
// NOTE: Avoid useState when initializing the query client if you don't
// have a suspense boundary between this and the code that may
// suspend because React will throw away the client on the initial
// render if it suspends and there is no boundary
const queryClient = getQueryClient();
const [trpcClient] = useState(() =>
createTRPCClient<AppRouter>({
links: [
httpBatchLink({
// transformer: superjson, <-- if you use a data transformer
url: getUrl(),
}),
],
}),
);
return (
<QueryClientProvider client={queryClient}>
<TRPCProvider trpcClient={trpcClient} queryClient={queryClient}>
{props.children}
</TRPCProvider>
</QueryClientProvider>
);
}Mount the provider in the root layout of your application:
import { TRPCReactProvider } from '~/trpc/client';
export default function RootLayout({
children,
}: Readonly<{
children: React.ReactNode;
}>) {
return (
<html lang="en">
<body>
<TRPCReactProvider>{children}</TRPCReactProvider>
</body>
</html>
);
}6. Create a tRPC caller for Server Components
To prefetch queries from server components, we create a proxy from our router. You can also pass in a client if your router is on a separate server.
import 'server-only'; // <-- ensure this file cannot be imported from the client
import { createTRPCOptionsProxy } from '@trpc/tanstack-react-query';
import { headers } from 'next/headers';
import { cache } from 'react';
import { createTRPCContext } from './init';
import { makeQueryClient } from './query-client';
import { appRouter } from './routers/_app';
// IMPORTANT: Create a stable getter for the query client that
// will return the same client during the same request.
export const getQueryClient = cache(makeQueryClient);
export const trpc = createTRPCOptionsProxy({
ctx: async () =>
createTRPCContext({
headers: await headers(),
}),
router: appRouter,
queryClient: getQueryClient,
});
// If your router is on a separate server, pass a client instead:
// createTRPCOptionsProxy({
// client: createTRPCClient({ links: [httpLink({ url: '...' })] }),
// queryClient: getQueryClient,
// });7. Make API requests
You're all set! You can now prefetch queries in server components and consume them in client components.
Prefetching in a Server Component
import { dehydrate, HydrationBoundary } from '@tanstack/react-query';
import { getQueryClient, trpc } from '~/trpc/server';
import { ClientGreeting } from './client-greeting';
export default async function Home() {
const queryClient = getQueryClient();
void queryClient.prefetchQuery(
trpc.hello.queryOptions({
text: 'world',
}),
);
return (
<HydrationBoundary state={dehydrate(queryClient)}>
<ClientGreeting />
</HydrationBoundary>
);
}Using data in a Client Component
'use client';
// <-- hooks can only be used in client components
import { useQuery } from '@tanstack/react-query';
import { useTRPC } from '~/trpc/client';
export function ClientGreeting() {
const trpc = useTRPC();
const greeting = useQuery(trpc.hello.queryOptions({ text: 'world' }));
if (!greeting.data) return <div>Loading...</div>;
return <div>{greeting.data.greeting}</div>;
}Leveraging Suspense
You may prefer handling loading and error states using Suspense and Error Boundaries. You can do this by using the useSuspenseQuery hook.
import { HydrateClient, prefetch, trpc } from '~/trpc/server';
import { Suspense } from 'react';
import { ErrorBoundary } from 'react-error-boundary';
import { ClientGreeting } from './client-greeting';
export default async function Home() {
prefetch(trpc.hello.queryOptions());
return (
<HydrateClient>
<ErrorBoundary fallback={<div>Something went wrong</div>}>
<Suspense fallback={<div>Loading...</div>}>
<ClientGreeting />
</Suspense>
</ErrorBoundary>
</HydrateClient>
);
}'use client';
import { useSuspenseQuery } from '@tanstack/react-query';
import { useTRPC } from '~/trpc/client';
export function ClientGreeting() {
const trpc = useTRPC();
const { data } = useSuspenseQuery(trpc.hello.queryOptions());
return <div>{data.greeting}</div>;
}Getting data in a Server Component
If you need access to the data in a server component, we recommend creating a server caller and using it directly. Please note that this method is detached from your query client and does not store the data in the cache. This means that you cannot use the data in a server component and expect it to be available in the client. This is intentional and explained in more detail in the Advanced Server Rendering guide.
import { headers } from 'next/headers';
import { createTRPCContext } from './init';
import { appRouter } from './routers/_app';
// ...
export const caller = appRouter.createCaller(async () =>
createTRPCContext({ headers: await headers() }),
);import { caller } from '~/trpc/server';
export default async function Home() {
const greeting = await caller.hello();
return <div>{greeting.greeting}</div>;
}If you really need to use the data both on the server as well as inside client components and understand the tradeoffs explained in the
Advanced Server Rendering
guide, you can use fetchQuery instead of prefetch to have the data both on the server as well as hydrating it down to the client:
import { getQueryClient, HydrateClient, trpc } from '~/trpc/server';
import { ClientGreeting } from './client-greeting';
export default async function Home() {
const queryClient = getQueryClient();
const greeting = await queryClient.fetchQuery(trpc.hello.queryOptions());
// Do something with greeting on the server
return (
<HydrateClient>
<ClientGreeting />
</HydrateClient>
);
}Next steps
- Learn about Server Actions for defining tRPC-powered server actions
- Learn about queries and mutations in client components
- Explore server-side calls for more advanced server-side patterns
- Check out the SSE Chat example for a full App Router example with subscriptions