[!IMPORTANT] We're actively working on a new alpha version that migrates from node-fetch to builtin fetch.

Please try it out and let us know if you run into any issues! https://github.com/anthropics/anthropic-sdk-typescript/issues/645

Anthropic TypeScript API Library

This library provides convenient access to the Anthropic REST API from server-side TypeScript or JavaScript.

The REST API documentation can be found on docs.anthropic.com. The full API of this library can be found in api.md.

Installation

1npm install @anthropic-ai/sdk

Usage

The full API of this library can be found in api.md.

1import Anthropic from '@anthropic-ai/sdk';
2
3const client = new Anthropic({
4  apiKey: process.env['ANTHROPIC_API_KEY'], // This is the default and can be omitted
5});
6
7async function main() {
8  const message = await client.messages.create({
9    max_tokens: 1024,
10    messages: [{ role: 'user', content: 'Hello, Claude' }],
11    model: 'claude-3-5-sonnet-latest',
12  });
13
14  console.log(message.content);
15}
16
17main();

Streaming responses

We provide support for streaming responses using Server Sent Events (SSE).

1import Anthropic from '@anthropic-ai/sdk';
2
3const client = new Anthropic();
4
5const stream = await client.messages.create({
6  max_tokens: 1024,
7  messages: [{ role: 'user', content: 'Hello, Claude' }],
8  model: 'claude-3-5-sonnet-latest',
9  stream: true,
10});
11for await (const messageStreamEvent of stream) {
12  console.log(messageStreamEvent.type);
13}

If you need to cancel a stream, you can break from the loop or call stream.controller.abort().

Request & Response types

This library includes TypeScript definitions for all request params and response fields. You may import and use them like so:

1import Anthropic from '@anthropic-ai/sdk';
2
3const client = new Anthropic({
4  apiKey: process.env['ANTHROPIC_API_KEY'], // This is the default and can be omitted
5});
6
7async function main() {
8  const params: Anthropic.MessageCreateParams = {
9    max_tokens: 1024,
10    messages: [{ role: 'user', content: 'Hello, Claude' }],
11    model: 'claude-3-5-sonnet-latest',
12  };
13  const message: Anthropic.Message = await client.messages.create(params);
14}
15
16main();

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

Counting Tokens

You can see the exact usage for a given request through the usage response property, e.g.

1const message = await client.messages.create(...)
2console.log(message.usage)
3// { input_tokens: 25, output_tokens: 13 }

Streaming Helpers

This library provides several conveniences for streaming messages, for example:

1import Anthropic from '@anthropic-ai/sdk';
2
3const anthropic = new Anthropic();
4
5async function main() {
6  const stream = anthropic.messages
7    .stream({
8      model: 'claude-3-5-sonnet-latest',
9      max_tokens: 1024,
10      messages: [
11        {
12          role: 'user',
13          content: 'Say hello there!',
14        },
15      ],
16    })
17    .on('text', (text) => {
18      console.log(text);
19    });
20
21  const message = await stream.finalMessage();
22  console.log(message);
23}
24
25main();

Streaming with client.messages.stream(...) exposes various helpers for your convenience including event handlers and accumulation.

Alternatively, you can use client.messages.create({ ..., stream: true }) which only returns an async iterable of the events in the stream and thus uses less memory (it does not build up a final message object for you).

Message Batches

This SDK provides beta support for the Message Batches API under the client.beta.messages.batches namespace.

Creating a batch

Message Batches takes an array of requests, where each object has a custom_id identifier, and the exact same request params as the standard Messages API:

1await anthropic.beta.messages.batches.create({
2  requests: [
3    {
4      custom_id: 'my-first-request',
5      params: {
6        model: 'claude-3-5-sonnet-latest',
7        max_tokens: 1024,
8        messages: [{ role: 'user', content: 'Hello, world' }],
9      },
10    },
11    {
12      custom_id: 'my-second-request',
13      params: {
14        model: 'claude-3-5-sonnet-latest',
15        max_tokens: 1024,
16        messages: [{ role: 'user', content: 'Hi again, friend' }],
17      },
18    },
19  ],
20});

Getting results from a batch

Once a Message Batch has been processed, indicated by .processing_status === 'ended', you can access the results with .batches.results()

1const results = await anthropic.beta.messages.batches.results(batch_id);
2for await (const entry of results) {
3  if (entry.result.type === 'succeeded') {
4    console.log(entry.result.message.content)
5  }
6}

Tool use beta

This SDK provides beta support for tool use, aka function calling. More details can be found in the documentation.

AWS Bedrock

We provide support for the Anthropic Bedrock API through a separate package.

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:

1async function main() {
2  const message = await client.messages
3    .create({
4      max_tokens: 1024,
5      messages: [{ role: 'user', content: 'Hello, Claude' }],
6      model: 'claude-3-5-sonnet-latest',
7    })
8    .catch(async (err) => {
9      if (err instanceof Anthropic.APIError) {
10        console.log(err.status); // 400
11        console.log(err.name); // BadRequestError
12        console.log(err.headers); // {server: 'nginx', ...}
13      } else {
14        throw err;
15      }
16    });
17}
18
19main();

Error codes are as followed:

Status Code	Error Type
400	BadRequestError
401	AuthenticationError
403	PermissionDeniedError
404	NotFoundError
422	UnprocessableEntityError
429	RateLimitError
>=500	InternalServerError
N/A	APIConnectionError

Request IDs

For more information on debugging requests, see these docs

All object responses in the SDK provide a _request_id property which is added from the request-id response header so that you can quickly log failing requests and report them back to Anthropic.

1const message = await client.messages.create({ max_tokens: 1024, messages: [{ role: 'user', content: 'Hello, Claude' }], model: 'claude-3-5-sonnet-latest' });
2console.log(message._request_id) // req_018EeWyXxfu5pfWkrYcMdjWG

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:

1// Configure the default for all requests:
2const client = new Anthropic({
3  maxRetries: 0, // default is 2
4});
5
6// Or, configure per-request:
7await client.messages.create({ max_tokens: 1024, messages: [{ role: 'user', content: 'Hello, Claude' }], model: 'claude-3-5-sonnet-latest' }, {
8  maxRetries: 5,
9});

Timeouts

By default requests time out after 10 minutes. However if you have specified a large max_tokens value and are not streaming, the default timeout will be calculated dynamically using the formula:

1const minimum = 10 * 60;
2const calculated = (60 * 60 * maxTokens) / 128_000;
3return calculated < minimum ? minimum * 1000 : calculated * 1000;

which will result in a timeout up to 60 minutes, scaled by the max_tokens parameter, unless overriden at the request or client level.

You can configure this with a timeout option:

1// Configure the default for all requests:
2const client = new Anthropic({
3  timeout: 20 * 1000, // 20 seconds (default is 10 minutes)
4});
5
6// Override per-request:
7await client.messages.create({ max_tokens: 1024, messages: [{ role: 'user', content: 'Hello, Claude' }], model: 'claude-3-5-sonnet-latest' }, {
8  timeout: 5 * 1000,
9});

On timeout, an APIConnectionTimeoutError is thrown.

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

Long Requests

[!IMPORTANT] We highly encourage you use the streaming Messages API for longer running requests.

We do not recommend setting a large max_tokens values without using streaming. Some networks may drop idle connections after a certain period of time, which can cause the request to fail or timeout without receiving a response from Anthropic.

This SDK will also throw an error if a non-streaming request is expected to be above roughly 10 minutes long. Passing stream: true or overriding the timeout option at the client or request level disables this error.

An expected request latency longer than the timeout for a non-streaming request will result in the client terminating the connection and retrying without receiving a response.

When supported by the fetch implementation, we set a TCP socket keep-alive option in order to reduce the impact of idle connection timeouts on some networks.

Auto-pagination

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

1async function fetchAllBetaMessagesBatches(params) {
2  const allBetaMessagesBatches = [];
3  // Automatically fetches more pages as needed.
4  for await (const betaMessageBatch of client.beta.messages.batches.list({ limit: 20 })) {
5    allBetaMessagesBatches.push(betaMessageBatch);
6  }
7  return allBetaMessagesBatches;
8}

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

1let page = await client.beta.messages.batches.list({ limit: 20 });
2for (const betaMessageBatch of page.data) {
3  console.log(betaMessageBatch);
4}
5
6// Convenience methods are provided for manually paginating:
7while (page.hasNextPage()) {
8  page = await page.getNextPage();
9  // ...
10}

Default Headers

We automatically send the anthropic-version header set to 2023-06-01.

If you need to, you can override it by setting default headers on a per-request basis.

Be aware that doing so may result in incorrect types and other unexpected or undefined behavior in the SDK.

1import Anthropic from '@anthropic-ai/sdk';
2
3const client = new Anthropic();
4
5const message = await client.messages.create(
6  {
7    max_tokens: 1024,
8    messages: [{ role: 'user', content: 'Hello, Claude' }],
9    model: 'claude-3-5-sonnet-latest',
10  },
11  { headers: { 'anthropic-version': 'My-Custom-Value' } },
12);

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.

You can also use the .withResponse() method to get the raw Response along with the parsed data.

1const client = new Anthropic();
2
3const response = await client.messages
4  .create({
5    max_tokens: 1024,
6    messages: [{ role: 'user', content: 'Hello, Claude' }],
7    model: 'claude-3-5-sonnet-latest',
8  })
9  .asResponse();
10console.log(response.headers.get('X-My-Header'));
11console.log(response.statusText); // access the underlying Response object
12
13const { data: message, response: raw } = await client.messages
14  .create({
15    max_tokens: 1024,
16    messages: [{ role: 'user', content: 'Hello, Claude' }],
17    model: 'claude-3-5-sonnet-latest',
18  })
19  .withResponse();
20console.log(raw.headers.get('X-My-Header'));
21console.log(message.content);

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 client.get, client.post, and other HTTP verbs. Options on the client, such as retries, will be respected when making these requests.

1await client.post('/some/path', {
2  body: { some_prop: 'foo' },
3  query: { some_query_arg: 'bar' },
4});

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.

1client.foo.create({
2  foo: 'my_param',
3  bar: 12,
4  // @ts-expect-error baz is not yet public
5  baz: 'undocumented option',
6});

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 uses node-fetch in Node, and expects a global fetch function in other environments.

If you would prefer to use a global, web-standards-compliant fetch function even in a Node environment, (for example, if you are running Node with --experimental-fetch or using NextJS which polyfills with undici), add the following import before your first import from "Anthropic":

1// Tell TypeScript and the package to use the global web fetch instead of node-fetch.
2// Note, despite the name, this does not add any polyfills, but expects them to be provided if needed.
3import '@anthropic-ai/sdk/shims/web';
4import Anthropic from '@anthropic-ai/sdk';

To do the inverse, add import "@anthropic-ai/sdk/shims/node" (which does import polyfills). This can also be useful if you are getting the wrong TypeScript types for Response (more details).

Logging and middleware

You may also provide a custom fetch function when instantiating the client, which can be used to inspect or alter the Request or Response before/after each request:

1import { fetch } from 'undici'; // as one example
2import Anthropic from '@anthropic-ai/sdk';
3
4const client = new Anthropic({
5  fetch: async (url: RequestInfo, init?: RequestInit): Promise<Response> => {
6    console.log('About to make a request', url, init);
7    const response = await fetch(url, init);
8    console.log('Got response', response);
9    return response;
10  },
11});

Note that if given a DEBUG=true environment variable, this library will log all requests and responses automatically. This is intended for debugging purposes only and may change in the future without notice.

Configuring an HTTP(S) Agent (e.g., for proxies)

By default, this library uses a stable agent for all http/https requests to reuse TCP connections, eliminating many TCP & TLS handshakes and shaving around 100ms off most requests.

If you would like to disable or customize this behavior, for example to use the API behind a proxy, you can pass an httpAgent which is used for all requests (be they http or https), for example:

1import http from 'http';
2import { HttpsProxyAgent } from 'https-proxy-agent';
3
4// Configure the default for all requests:
5const client = new Anthropic({
6  httpAgent: new HttpsProxyAgent(process.env.PROXY_URL),
7});
8
9// Override per-request:
10await client.messages.create(
11  {
12    max_tokens: 1024,
13    messages: [{ role: 'user', content: 'Hello, Claude' }],
14    model: 'claude-3-5-sonnet-latest',
15  },
16  {
17    httpAgent: new http.Agent({ keepAlive: false }),
18  },
19);

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 >= 4.5 is supported.

The following runtimes are supported:

• Node.js 18 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.
• Web browsers: disabled by default to avoid exposing your secret API credentials (see our help center for best practices). Enable browser support by explicitly setting dangerouslyAllowBrowser to true.

More explanation

Why is this dangerous?

Enabling the dangerouslyAllowBrowser option can be dangerous because it exposes your secret API credentials in the client-side code. Web browsers are inherently less secure than server environments, any user with access to the browser can potentially inspect, extract, and misuse these credentials. This could lead to unauthorized access using your credentials and potentially compromise sensitive data or functionality.

When might this not be dangerous?

In certain scenarios where enabling browser support might not pose significant risks:

• Internal Tools: If the application is used solely within a controlled internal environment where the users are trusted, the risk of credential exposure can be mitigated.
• Development or debugging purpose: Enabling this feature temporarily might be acceptable, provided the credentials are short-lived, aren't also used in production environments, or are frequently rotated.

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.