Replicate TypeScript SDK

This library provides an npm package for using Replicate's HTTP API from server-side TypeScript or JavaScript.

• HTTP API docs: replicate.com/docs/reference/http
• TypeScript SDK docs: sdks.replicate.com/typescript

Installation

npm install replicate@alpha

Usage

Find the full API reference for this SDK at sdks.replicate.com/typescript.

import Replicate from 'replicate';

const replicate = new Replicate({
  bearerToken: process.env['REPLICATE_API_TOKEN'], // This is the default and can be omitted
});

const prediction = await replicate.predictions.get({ prediction_id: 'gm3qorzdhgbfurvjtvhg6dckhu' });

console.log(prediction.id);

The SDK exposes different layers of abstraction depending on what you are trying to achieve.

Run a model — replicate.run()

import Replicate from 'replicate-stainless';

const replicate = new Replicate();
const model = 'anthropic/claude-4-sonnet';
const input = { prompt: 'Write a poem about a cat named Claude.' };
const output = await replicate.run(model, { input });
console.log(output.join(''));

To run a model with a file input, you can pass an HTTPS URL to a publicly accessible file or a file handle from your device.

Using HTTPS URLs as file inputs

Here's an example of how to run a model using an HTTPS URL as a file input:

import { Replicate } from 'replicate-stainless';

const replicate = new Replicate();
const model = 'black-forest-labs/flux-kontext-pro';
const input = {
  input_image:
    'https://replicate.delivery/pbxt/N5cepICxyaagdvULl0phi7ImdxuFz05TR2l623zqxhNR9q5Y/van-gogh.jpeg',
  prompt: 'Using this style, a frog riding on a penny-farthing bicycle',
};
const output = await replicate.run(model, { input });

console.log(output);

Using a local file handle as input

Here's an example of how to run a model with a local file handle as a file input:

import fs from 'node:fs/promises';
import { Replicate } from 'replicate-stainless';

const replicate = new Replicate();
const model = 'black-forest-labs/flux-kontext-pro';
const input = {
  input_image: await fs.readFile('sweater.png'),
  prompt: 'make it a 90s cartoon style',
};
const output = await replicate.run(model, { input });

console.log(output);

File handle inputs are automatically uploaded to Replicate. For larger files (>100MiB), upload the file to your own storage provider and pass a publicly accessible URL.

replicate.run raises ModelError if the prediction fails. You can access the exception's prediction property to get more information about the failure.

Resource API — replicate.predictions.*, replicate.models.*, …

The full API, including all definitions for all request params and responses, is exposed in the client. The full list of methods can be found in api.md.

You may import and use them like so:

import Replicate from 'replicate';

const replicate = new Replicate({
  bearerToken: process.env['REPLICATE_API_TOKEN'], // This is the default and can be omitted
});

const params: Replicate.PredictionCreateParams = {
  input: { text: 'Alice' },
  version: 'replicate/hello-world:5c7d5dc6dd8bf75c1acaa8565735e7986bc5b66206b55cca93cb72c9bf15ccaa',
};
const prediction: Replicate.Prediction = await replicate.predictions.create(params);

const collections = replicate.collections.list();

const accounts = await replicate.accounts.list();

Documentation for each method, request param, and response field are available in docstrings and will appear on hover in most modern editors:

Low‑level requests — replicate.get() / replicate.post() / …

The SDK also exposes methods to make generic requests to the API. This is helpful for handling experimental or undocumented endpoints, or other non-standard requests.

await replicate.post('/v1/experimental/feature', {
  body: { foo: 'bar' },
  query: { alpha: true },
});

File uploads

Request parameters that correspond to file uploads can be passed in many different forms:

• File (or an object with the same structure)
• a fetch Response (or an object with the same structure)
• an fs.ReadStream
• the return value of our toFile helper

import fs from 'fs';
import Replicate, { toFile } from 'replicate';

const replicate = new Replicate();

// If you have access to Node `fs` we recommend using `fs.createReadStream()`:
await replicate.files.create({ content: fs.createReadStream('/path/to/file') });

// Or if you have the web `File` API you can pass a `File` instance:
await replicate.files.create({ content: new File(['my bytes'], 'file') });

// You can also pass a `fetch` `Response`:
await replicate.files.create({ content: await fetch('https://somesite/file') });

// Finally, if none of the above are convenient, you can use our `toFile` helper:
await replicate.files.create({ content: await toFile(Buffer.from('my bytes'), 'file') });
await replicate.files.create({ content: await toFile(new Uint8Array([0, 1, 2]), 'file') });

Verifying webhooks

To prevent unauthorized requests, Replicate signs every webhook and its metadata with a unique key for each user or organization. The helper validateWebhook can be used to verify the webhook indeed comes from Replicate before processing it.

import { validateWebhook } from 'replicate';

const secret = await process.env['REPLICATE_WEBHOOK_SIGNING_SECRET'];

const replicate = new Replicate();

// In your webhook handler (e.g., Next.js API route)
export async function POST(request: Request) {
  const isValid = await validateWebhook(request.clone(), secret);
  if (!isValid) {
    return new Response('Invalid webhook signature', { status: 401 });
  }

  const body = await request.json();
  console.log('Prediction completed:', body);

  return new Response('OK', { status: 200 });
}

For environments that don't use the Fetch API Request object, you can pass the webhook parameters directly:

import { validateWebhook } from 'replicate';

const isValid = await validateWebhook({
  id: webhookHeaders['webhook-id'],
  timestamp: webhookHeaders['webhook-timestamp'],
  signature: webhookHeaders['webhook-signature'],
  body: webhookBody,
  secret: webhookSecret,
});

Note for Node.js 18: You may need to pass the crypto implementation:

import { webcrypto } from 'node:crypto';

const isValid = await validateWebhook(request, secret.key, webcrypto);

Handling errors

When the library is unable to connect to the API, or if the API returns a non-success status code (i.e., 4xx or 5xx response), a subclass of APIError will be thrown:

const prediction = await replicate.predictions
  .create({
    input: { text: 'Alice' },
    version: 'replicate/hello-world:5c7d5dc6dd8bf75c1acaa8565735e7986bc5b66206b55cca93cb72c9bf15ccaa',
  })
  .catch(async (err) => {
    if (err instanceof Replicate.APIError) {
      console.log(err.status); // 400
      console.log(err.name); // BadRequestError
      console.log(err.headers); // {server: 'nginx', ...}
    } else {
      throw err;
    }
  });

Error codes are as follows:

Retries

Certain errors will be automatically retried 2 times by default, with a short exponential backoff. Connection errors (for example, due to a network connectivity problem), 408 Request Timeout, 409 Conflict, 429 Rate Limit, and >=500 Internal errors will all be retried by default.

You can use the maxRetries option to configure or disable this:

// Configure the default for all requests:
const replicate = new Replicate({
  maxRetries: 0, // default is 2
});

// Or, configure per-request:
await replicate.predictions.create({ input: { text: 'Alice' }, version: 'replicate/hello-world:5c7d5dc6dd8bf75c1acaa8565735e7986bc5b66206b55cca93cb72c9bf15ccaa' }, {
  maxRetries: 5,
});

Timeouts

Requests time out after 1 minute by default. You can configure this with a timeout option:

// Configure the default for all requests:
const replicate = new Replicate({
  timeout: 20 * 1000, // 20 seconds (default is 1 minute)
});

// Override per-request:
await replicate.predictions.create({ input: { text: 'Alice' }, version: 'replicate/hello-world:5c7d5dc6dd8bf75c1acaa8565735e7986bc5b66206b55cca93cb72c9bf15ccaa' }, {
  timeout: 5 * 1000,
});

On timeout, an APIConnectionTimeoutError is thrown.

Note that requests which time out will be retried twice by default.

Auto-pagination

List methods in the Replicate API are paginated. You can use the for await … of syntax to iterate through items across all pages:

async function fetchAllModelListResponses(params) {
  const allModelListResponses = [];
  // Automatically fetches more pages as needed.
  for await (const modelListResponse of replicate.models.list()) {
    allModelListResponses.push(modelListResponse);
  }
  return allModelListResponses;
}

Alternatively, you can request a single page at a time:

let page = await replicate.models.list();
for (const modelListResponse of page.results) {
  console.log(modelListResponse);
}

// Convenience methods are provided for manually paginating:
while (page.hasNextPage()) {
  page = await page.getNextPage();
  // ...
}

Advanced Usage

Accessing raw Response data (e.g., headers)

The "raw" Response returned by fetch() can be accessed through the .asResponse() method on the APIPromise type that all methods return. This method returns as soon as the headers for a successful response are received and does not consume the response body, so you are free to write custom parsing or streaming logic.

You can also use the .withResponse() method to get the raw Response along with the parsed data. Unlike .asResponse() this method consumes the body, returning once it is parsed.

const replicate = new Replicate();

const response = await replicate.predictions
  .create({
    input: { text: 'Alice' },
    version: 'replicate/hello-world:5c7d5dc6dd8bf75c1acaa8565735e7986bc5b66206b55cca93cb72c9bf15ccaa',
  })
  .asResponse();
console.log(response.headers.get('X-My-Header'));
console.log(response.statusText); // access the underlying Response object

const { data: prediction, response: raw } = await replicate.predictions
  .create({
    input: { text: 'Alice' },
    version: 'replicate/hello-world:5c7d5dc6dd8bf75c1acaa8565735e7986bc5b66206b55cca93cb72c9bf15ccaa',
  })
  .withResponse();
console.log(raw.headers.get('X-My-Header'));
console.log(prediction.id);

Logging

[!IMPORTANT] All log messages are intended for debugging only. The format and content of log messages may change between releases.

Log levels

The log level can be configured in two ways:

1. Via the REPLICATE_LOG environment variable
2. Using the logLevel client option (overrides the environment variable if set)

import Replicate from 'replicate';

const replicate = new Replicate({
  logLevel: 'debug', // Show all log messages
});

Available log levels, from most to least verbose:

• 'debug' - Show debug messages, info, warnings, and errors
• 'info' - Show info messages, warnings, and errors
• 'warn' - Show warnings and errors (default)
• 'error' - Show only errors
• 'off' - Disable all logging

At the 'debug' level, all HTTP requests and responses are logged, including headers and bodies. Some authentication-related headers are redacted, but sensitive data in request and response bodies may still be visible.

Custom logger

By default, this library logs to globalThis.console. You can also provide a custom logger. Most logging libraries are supported, including pino, winston, bunyan, consola, signale, and @std/log. If your logger doesn't work, please open an issue.

When providing a custom logger, the logLevel option still controls which messages are emitted, messages below the configured level will not be sent to your logger.

import Replicate from 'replicate';
import pino from 'pino';

const logger = pino();

const replicate = new Replicate({
  logger: logger.child({ name: 'Replicate' }),
  logLevel: 'debug', // Send all messages to pino, allowing it to filter
});

Making custom/undocumented requests

This library is typed for convenient access to the documented API. If you need to access undocumented endpoints, params, or response properties, the library can still be used.

Undocumented endpoints

To make requests to undocumented endpoints, you can use replicate.get, replicate.post, and other HTTP verbs. Options on the client, such as retries, will be respected when making these requests.

await replicate.post('/some/path', {
  body: { some_prop: 'foo' },
  query: { some_query_arg: 'bar' },
});

Undocumented request params

To make requests using undocumented parameters, you may use // @ts-expect-error on the undocumented parameter. This library doesn't validate at runtime that the request matches the type, so any extra values you send will be sent as-is.

replicate.predictions.get({
  // ...
  // @ts-expect-error baz is not yet public
  baz: 'undocumented option',
});

For requests with the GET verb, any extra params will be in the query, all other requests will send the extra param in the body.

If you want to explicitly send an extra argument, you can do so with the query, body, and headers request options.

Undocumented response properties

To access undocumented response properties, you may access the response object with // @ts-expect-error on the response object, or cast the response object to the requisite type. Like the request params, we do not validate or strip extra properties from the response from the API.

Customizing the fetch client

By default, this library expects a global fetch function is defined.

If you want to use a different fetch function, you can either polyfill the global:

import fetch from 'my-fetch';

globalThis.fetch = fetch;

Or pass it to the client:

import Replicate from 'replicate';
import fetch from 'my-fetch';

const replicate = new Replicate({ fetch });

Fetch options

If you want to set custom fetch options without overriding the fetch function, you can provide a fetchOptions object when instantiating the client or making a request. (Request-specific options override client options.)

import Replicate from 'replicate';

const replicate = new Replicate({
  fetchOptions: {
    // `RequestInit` options
  },
});

Configuring proxies

To modify proxy behavior, you can provide custom fetchOptions that add runtime-specific proxy options to requests:

Node ^[docs]

import Replicate from 'replicate';
import * as undici from 'undici';

const proxyAgent = new undici.ProxyAgent('http://localhost:8888');
const replicate = new Replicate({
  fetchOptions: {
    dispatcher: proxyAgent,
  },
});

Bun ^[docs]

import Replicate from 'replicate';

const replicate = new Replicate({
  fetchOptions: {
    proxy: 'http://localhost:8888',
  },
});

Deno ^[docs]

import Replicate from 'npm:replicate';

const httpClient = Deno.createHttpClient({ proxy: { url: 'http://localhost:8888' } });
const replicate = new Replicate({
  fetchOptions: {
    client: httpClient,
  },
});

Frequently Asked Questions

Semantic versioning

This package generally follows SemVer conventions, though certain backwards-incompatible changes may be released as minor versions:

1. Changes that only affect static types, without breaking runtime behavior.
2. Changes to library internals which are technically public but not intended or documented for external use. (Please open a GitHub issue to let us know if you are relying on such internals.)
3. Changes that we do not expect to impact the vast majority of users in practice.

We take backwards-compatibility seriously and work hard to ensure you can rely on a smooth upgrade experience.

We are keen for your feedback; please open an issue with questions, bugs, or suggestions.

Requirements

TypeScript >= 4.9 is supported.

The following runtimes are supported:

• Web browsers (Up-to-date Chrome, Firefox, Safari, Edge, and more)
• Node.js 20 LTS or later (non-EOL) versions.
• Deno v1.28.0 or higher.
• Bun 1.0 or later.
• Cloudflare Workers.
• Vercel Edge Runtime.
• Jest 28 or greater with the "node" environment ("jsdom" is not supported at this time).
• Nitro v2.6 or greater.

Note that React Native is not supported at this time.

If you are interested in other runtime environments, please open or upvote an issue on GitHub.

Contributing

See the contributing documentation. ✌️