OpenAI TypeScript and JavaScript API Library

This library provides convenient access to the OpenAI REST API from TypeScript or JavaScript.

It is generated from our OpenAPI specification with Stainless.

To learn how to use the OpenAI API, check out our API Reference and Documentation.

Installation

npm install openai

Installation from JSR

deno add jsr:@openai/openai
npx jsr add @openai/openai

These commands will make the module importable from the @openai/openai scope. You can also import directly from JSR without an install step if you're using the Deno JavaScript runtime:

import OpenAI from 'jsr:@openai/openai';

Usage

The full API of this library can be found in api.md file along with many code examples.

The primary API for interacting with OpenAI models is the Responses API. You can generate text from the model with the code below.

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env['OPENAI_API_KEY'], // This is the default and can be omitted
});

const response = await client.responses.create({
  model: 'gpt-4o',
  instructions: 'You are a coding assistant that talks like a pirate',
  input: 'Are semicolons optional in JavaScript?',
});

console.log(response.output_text);

The previous standard (supported indefinitely) for generating text is the Chat Completions API. You can use that API to generate text from the model with the code below.

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env['OPENAI_API_KEY'], // This is the default and can be omitted
});

const completion = await client.chat.completions.create({
  model: 'gpt-4o',
  messages: [
    { role: 'developer', content: 'Talk like a pirate.' },
    { role: 'user', content: 'Are semicolons optional in JavaScript?' },
  ],
});

console.log(completion.choices[0].message.content);

Streaming responses

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

import OpenAI from 'openai';

const client = new OpenAI();

const stream = await client.responses.create({
  model: 'gpt-4o',
  input: 'Say "Sheep sleep deep" ten times fast!',
  stream: true,
});

for await (const event of stream) {
  console.log(event);
}

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 fetch from 'node-fetch';
import OpenAI, { toFile } from 'openai';

const client = new OpenAI();

// If you have access to Node `fs` we recommend using `fs.createReadStream()`:
await client.files.create({ file: fs.createReadStream('input.jsonl'), purpose: 'fine-tune' });

// Or if you have the web `File` API you can pass a `File` instance:
await client.files.create({ file: new File(['my bytes'], 'input.jsonl'), purpose: 'fine-tune' });

// You can also pass a `fetch` `Response`:
await client.files.create({ file: await fetch('https://somesite/input.jsonl'), purpose: 'fine-tune' });

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

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:

async function main() {
  const job = await client.fineTuning.jobs
    .create({ model: 'gpt-4o', training_file: 'file-abc123' })
    .catch(async (err) => {
      if (err instanceof OpenAI.APIError) {
        console.log(err.request_id);
        console.log(err.status); // 400
        console.log(err.name); // BadRequestError
        console.log(err.headers); // {server: 'nginx', ...}
      } else {
        throw err;
      }
    });
}

main();

Error codes are as followed:

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 client = new OpenAI({
  maxRetries: 0, // default is 2
});

// Or, configure per-request:
await client.chat.completions.create({ messages: [{ role: 'user', content: 'How can I get the name of the current day in JavaScript?' }], model: 'gpt-4o' }, {
  maxRetries: 5,
});

Timeouts

Requests time out after 10 minutes by default. You can configure this with a timeout option:

// Configure the default for all requests:
const client = new OpenAI({
  timeout: 20 * 1000, // 20 seconds (default is 10 minutes)
});

// Override per-request:
await client.chat.completions.create({ messages: [{ role: 'user', content: 'How can I list all files in a directory using Python?' }], model: 'gpt-4o' }, {
  timeout: 5 * 1000,
});

On timeout, an APIConnectionTimeoutError is thrown.

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

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 x-request-id response header so that you can quickly log failing requests and report them back to OpenAI.

const response = await client.responses.create({ model: 'gpt-4o', input: 'testing 123' });
console.log(response._request_id) // req_123

You can also access the Request ID using the .withResponse() method:

const { data: stream, request_id } = await openai.responses
  .create({
    model: 'gpt-4o',
    input: 'Say this is a test',
    stream: true,
  })
  .withResponse();

Auto-pagination

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

async function fetchAllFineTuningJobs(params) {
  const allFineTuningJobs = [];
  // Automatically fetches more pages as needed.
  for await (const fineTuningJob of client.fineTuning.jobs.list({ limit: 20 })) {
    allFineTuningJobs.push(fineTuningJob);
  }
  return allFineTuningJobs;
}

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

let page = await client.fineTuning.jobs.list({ limit: 20 });
for (const fineTuningJob of page.data) {
  console.log(fineTuningJob);
}

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

Realtime API Beta

The Realtime API enables you to build low-latency, multi-modal conversational experiences. It currently supports text and audio as both input and output, as well as function calling through a WebSocket connection.

import { OpenAIRealtimeWebSocket } from 'openai/beta/realtime/websocket';

const rt = new OpenAIRealtimeWebSocket({ model: 'gpt-4o-realtime-preview-2024-12-17' });

rt.on('response.text.delta', (event) => process.stdout.write(event.delta));

For more information see realtime.md.

Microsoft Azure OpenAI

To use this library with Azure OpenAI, use the AzureOpenAI class instead of the OpenAI class.

[!IMPORTANT] The Azure API shape slightly differs from the core API shape which means that the static types for responses / params won't always be correct.

import { AzureOpenAI } from 'openai';
import { getBearerTokenProvider, DefaultAzureCredential } from '@azure/identity';

const credential = new DefaultAzureCredential();
const scope = 'https://cognitiveservices.azure.com/.default';
const azureADTokenProvider = getBearerTokenProvider(credential, scope);

const openai = new AzureOpenAI({ azureADTokenProvider, apiVersion: "<The API version, e.g. 2024-10-01-preview>" });

const result = await openai.chat.completions.create({
  model: 'gpt-4o',
  messages: [{ role: 'user', content: 'Say hello!' }],
});

console.log(result.choices[0]!.message?.content);

For more information on support for the Azure API, see azure.md.

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.

const client = new OpenAI();

const httpResponse = await client.responses
  .create({ model: 'gpt-4o', input: 'say this is a test.' })
  .asResponse();

// access the underlying web standard Response object
console.log(httpResponse.headers.get('X-My-Header'));
console.log(httpResponse.statusText);

const { data: modelResponse, response: raw } = await client.responses
  .create({ model: 'gpt-4o', input: 'say this is a test.' })
  .withResponse();
console.log(raw.headers.get('X-My-Header'));
console.log(modelResponse);

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.

await client.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.

client.foo.create({
  foo: 'my_param',
  bar: 12,
  // @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

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://community.openai.com/t/your-feedback-requested-node-js-sdk-5-0-0-alpha/1063774

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 "OpenAI":

// Tell TypeScript and the package to use the global web fetch instead of node-fetch.
// Note, despite the name, this does not add any polyfills, but expects them to be provided if needed.
import 'openai/shims/web';
import OpenAI from 'openai';

To do the inverse, add import "openai/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:

import { fetch } from 'undici'; // as one example
import OpenAI from 'openai';

const client = new OpenAI({
  fetch: async (url: RequestInfo, init?: RequestInit): Promise<Response> => {
    console.log('About to make a request', url, init);
    const response = await fetch(url, init);
    console.log('Got response', response);
    return response;
  },
});

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:

import http from 'http';
import { HttpsProxyAgent } from 'https-proxy-agent';

// Configure the default for all requests:
const client = new OpenAI({
  httpAgent: new HttpsProxyAgent(process.env.PROXY_URL),
});

// Override per-request:
await client.models.list({
  httpAgent: new http.Agent({ keepAlive: false }),
});

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. Enable browser support by explicitly setting dangerouslyAllowBrowser to true'.

  <summary>More explanation</summary>

  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.
  • Public APIs with Limited Scope: If your API has very limited scope and the exposed credentials do not grant access to sensitive data or critical operations, the potential impact of exposure is reduced.
  • 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.

Changelog

4.95.1 (2025-04-18)

Full Changelog: v4.95.0...v4.95.1

Bug Fixes

• zod: warn on optional field usage (#1469) (aea2d12)

4.95.0 (2025-04-16)

Full Changelog: v4.94.0...v4.95.0

Features

• api: add o3 and o4-mini model IDs (4845cd9)

4.94.0 (2025-04-14)

Full Changelog: v4.93.0...v4.94.0

Features

• api: adding gpt-4.1 family of model IDs (bddcbcf)
• api: manual updates (7532f48)

Chores

• client: minor internal fixes (d342f17)
• internal: reduce CI branch coverage (a49b94a)
• internal: upload builds and expand CI branch coverage (#1460) (7e23bb4)
• workaround build errors (913eba8)

4.93.0 (2025-04-08)

Full Changelog: v4.92.1...v4.93.0

Features

• api: Add evalapi to sdk (#1456) (ee917e3)

Chores

• internal: fix examples (#1457) (a3dd0dd)
• internal: skip broken test (#1458) (4d2f815)
• tests: improve enum examples (#1454) (ecabce2)

4.92.1 (2025-04-07)

Full Changelog: v4.92.0...v4.92.1

Chores

• internal: only run examples workflow in main repo (#1450) (5e49a7a)

4.92.0 (2025-04-07)

Full Changelog: v4.91.1...v4.92.0

Features

• api: manual updates (891754d)
• api: manual updates (01e5546)
• api: manual updates (f38dbf3)
• api: manual updates (1f12253)

Bug Fixes

• api: improve type resolution when importing as a package (#1444) (4aa46d6)
• client: send X-Stainless-Timeout in seconds (#1442) (aa4206c)
• embeddings: correctly decode base64 data (#1448) (58128f7)
• mcp: remove unused tools.ts (#1445) (520a8fa)

Chores

• internal: add aliases for Record and Array (#1443) (b65391b)

4.91.1 (2025-04-01)

Full Changelog: v4.91.0...v4.91.1

Bug Fixes

• docs: correct docstring on responses.stream (1c8cd6a)

Chores

• Remove deprecated/unused remote spec feature (ce3dfa8)

4.91.0 (2025-03-31)

Full Changelog: v4.90.0...v4.91.0

Features

• api: add get /responses/{response_id}/input_items endpoint (ef0e0ac)

Performance Improvements

• embedding: default embedding creation to base64 (#1312) (e54530e), closes #1310

4.90.0 (2025-03-27)

Full Changelog: v4.89.1...v4.90.0

Features

• api: add get /chat/completions endpoint (2d6710a)

Bug Fixes

• audio: correctly handle transcription streaming (2a9b603)
• internal: work around https://github.com/vercel/next.js/issues/76881 (#1427) (b467e94)

Chores

• add hash of OpenAPI spec/config inputs to .stats.yml (45db35e)
• api: updates to supported Voice IDs (#1424) (404f4db)
• client: expose headers on some streaming errors (#1423) (b0783cc)

4.89.1 (2025-03-26)

Full Changelog: v4.89.0...v4.89.1

Bug Fixes

• avoid type error in certain environments (#1413) (d3f6f8f)
• client: remove duplicate types (#1410) (338878b)
• exports: add missing type exports (#1417) (2d15ada)

Chores

• internal: version bump (#1408) (9c0949a)

4.89.0 (2025-03-20)

Full Changelog: v4.88.0...v4.89.0

Features

• add audio helpers (ea1b6b4)
• api: new models for TTS, STT, + new audio features for Realtime (#1407) (142933a)

Chores

• internal: version bump (#1400) (6838ab4)

4.88.0 (2025-03-19)

Full Changelog: v4.87.4...v4.88.0

Features

• api: o1-pro now available through the API (#1398) (616a7e9)

Chores

• exports: cleaner resource index imports (#1396) (26b0856)
• exports: stop using path fallbacks (#1397) (d1479c2)
• internal: version bump (#1393) (7f16c3a)

4.87.4 (2025-03-18)

Full Changelog: v4.87.3...v4.87.4

Bug Fixes

• api: correct some Responses types (#1391) (af45876)
• types: ignore missing id in responses pagination (1b9d20e)
• types: improve responses type names (#1392) (164f476)

Chores

• add missing type alias exports (#1390) (16c5e22)
• internal: add back release workflow (dddf29b)
• internal: remove CI condition (#1381) (ef17981)
• internal: run CI on update-specs branch (9fc2130)
• internal: update release workflows (90b77d0)

4.87.3 (2025-03-11)

Full Changelog: v4.87.2...v4.87.3

Bug Fixes

• responses: correct reasoning output type (2abef57)

4.87.2 (2025-03-11)

Full Changelog: v4.87.1...v4.87.2

Bug Fixes

• responses: correctly add output_text (4ceb5cc)

4.87.1 (2025-03-11)

Full Changelog: v4.87.0...v4.87.1

Bug Fixes

• correct imports (5cdf17c)

4.87.0 (2025-03-11)

Full Changelog: v4.86.2...v4.87.0

Features

• api: add /v1/responses and built-in tools (119b584)

4.86.2 (2025-03-05)

Full Changelog: v4.86.1...v4.86.2

Chores

• internal: run example files in CI (#1357) (88d0050)

4.86.1 (2025-02-27)

Full Changelog: v4.86.0...v4.86.1

Documentation

• update URLs from stainlessapi.com to stainless.com (#1352) (8294e9e)

4.86.0 (2025-02-27)

Full Changelog: v4.85.4...v4.86.0

Features

• api: add gpt-4.5-preview (#1349) (2a1d36b)

4.85.4 (2025-02-22)

Full Changelog: v4.85.3...v4.85.4

Chores

• internal: fix devcontainers setup (#1343) (cb1ec90)

4.85.3 (2025-02-20)

Full Changelog: v4.85.2...v4.85.3

Bug Fixes

• parsing: remove tool_calls default empty array (#1341) (2672160)

4.85.2 (2025-02-18)

Full Changelog: v4.85.1...v4.85.2

Bug Fixes

• optimize sse chunk reading off-by-one error (#1339) (c82795b)

4.85.1 (2025-02-14)

Full Changelog: v4.85.0...v4.85.1

Bug Fixes

• client: fix export map for index exports (#1328) (647ba7a)
• package: add chat/completions.ts back in (#1333) (e4b5546)

Chores

• internal: add missing return type annotation (#1334) (53e0856)

4.85.0 (2025-02-13)

Full Changelog: v4.84.1...v4.85.0

Features

• api: add support for storing chat completions (#1327) (8d77f8e)

Bug Fixes

• realtime: call .toString() on WebSocket url (#1324) (09bc50d)

4.84.1 (2025-02-13)

Full Changelog: v4.84.0...v4.84.1

Bug Fixes

• realtime: correct websocket type var constraint (#1321) (afb17ea)

4.84.0 (2025-02-12)

Full Changelog: v4.83.0...v4.84.0

Features

• pagination: avoid fetching when has_more: false (#1305) (b6944c6)

Bug Fixes

• api: add missing reasoning effort + model enums (#1302) (14c55c3)
• assistants: handle thread.run.incomplete event (7032cc4)
• correctly decode multi-byte characters over multiple chunks (#1316) (dd776c4)

Chores

• internal: remove segfault-handler dependency (3521ca3)

Documentation

• readme: cleanup into multiple files (da94424)

4.83.0 (2025-02-05)

Full Changelog: v4.82.0...v4.83.0

Features

• client: send X-Stainless-Timeout header (#1299) (ddfc686)

Bug Fixes

• api/types: correct audio duration & role types (#1300) (a955ac2)
• azure/audio: use model param for deployments (#1297) (85de382)

4.82.0 (2025-01-31)

Full Changelog: v4.81.0...v4.82.0

Features

• api: add o3-mini (#1295) (378e2f7)

Bug Fixes

• examples/realtime: remove duplicate session.update call (#1293) (ad800b4)
• types: correct metadata type + other fixes (378e2f7)

4.81.0 (2025-01-29)

Full Changelog: v4.80.1...v4.81.0

Features

• azure: Realtime API support (#1287) (fe090c0)

4.80.1 (2025-01-24)

Full Changelog: v4.80.0...v4.80.1

Bug Fixes

• azure: include retry count header (3e0ba40)

Documentation

• fix typo, "zodFunctionTool" -> "zodFunction" (#1128) (b7ab6bb)
• helpers: fix type annotation (fc019df)
• readme: fix realtime errors docs link (#1286) (d1d50c8)

4.80.0 (2025-01-22)

Full Changelog: v4.79.4...v4.80.0

Features

• api: update enum values, comments, and examples (#1280) (d38f2c2)

4.79.4 (2025-01-21)

Full Changelog: v4.79.3...v4.79.4

Bug Fixes

• jsr: correct zod config (e45fa5f)

Chores

• internal: minor restructuring (#1278) (58ea92a)

Documentation

• update deprecation messages (#1275) (1c6599e)

4.79.3 (2025-01-21)

Full Changelog: v4.79.2...v4.79.3

Bug Fixes

• jsr: export zod helpers (9dc55b6)

4.79.2 (2025-01-21)

Full Changelog: v4.79.1...v4.79.2

Chores

• internal: add test (#1270) (b7c2d3d)

Documentation

• readme: fix Realtime API example link (#1272) (d0653c7)

4.79.1 (2025-01-17)

Full Changelog: v4.79.0...v4.79.1

Bug Fixes

• realtime: correct import syntax (#1267) (74702a7)

4.79.0 (2025-01-17)

Full Changelog: v4.78.1...v4.79.0

Features

• client: add Realtime API support (#1266) (7160ebe)

Bug Fixes

• logs/azure: redact sensitive header when DEBUG is set (#1218) (6a72fd7)

Chores

• fix streaming (379c743)
• internal: streaming refactors (#1261) (dd4af93)
• types: add | undefined to client options properties (#1264) (5e56979)
• types: rename vector store chunking strategy (#1263) (d31acee)

4.78.1 (2025-01-10)

Full Changelog: v4.78.0...v4.78.1

Bug Fixes

• send correct Accept header for certain endpoints (#1257) (8756693)

4.78.0 (2025-01-09)

Full Changelog: v4.77.4...v4.78.0

Features

• client: add realtime types (#1254) (7130995)

4.77.4 (2025-01-08)

Full Changelog: v4.77.3...v4.77.4

Documentation

• readme: fix misplaced period (#1252) (c2fe465)

4.77.3 (2025-01-03)

Full Changelog: v4.77.2...v4.77.3

Chores

• api: bump spec version (#1248) (37b3df9)

4.77.2 (2025-01-02)

Full Changelog: v4.77.1...v4.77.2

Chores

• bump license year (#1246) (13197c1)

4.77.1 (2024-12-21)

Full Changelog: v4.77.0...v4.77.1

Bug Fixes

• client: normalize method (#1235) (4a213da)

Chores

• internal: spec update (#1231) (a97ea73)

Documentation

• minor formatting changes (#1236) (6387968)
• readme: add alpha callout (f2eff37)

4.77.0 (2024-12-17)

Full Changelog: v4.76.3...v4.77.0

Features

• api: new o1 and GPT-4o models + preference fine-tuning (#1229) (2e872d4)

Chores

• internal: fix some typos (#1227) (d51fcfe)
• internal: spec update (#1230) (ed2b61d)

4.76.3 (2024-12-13)

Full Changelog: v4.76.2...v4.76.3

Chores

• internal: better ecosystem test debugging (86fc0a8)

Documentation

• README: fix helpers section links (#1224) (efbe30a)

4.76.2 (2024-12-12)

Full Changelog: v4.76.1...v4.76.2

Chores

• internal: update isAbsoluteURL (#1223) (e908ed7)
• types: nicer error class types + jsdocs (#1219) (576d24c)

4.76.1 (2024-12-10)

Full Changelog: v4.76.0...v4.76.1

Chores

• internal: bump cross-spawn to v7.0.6 (#1217) (c07ad29)
• internal: remove unnecessary getRequestClient function (#1215) (bef3925)

4.76.0 (2024-12-05)

Full Changelog: v4.75.0...v4.76.0

Features

• api: updates (#1212) (e0fedf2)

Chores

• bump openapi url (#1210) (3fa95a4)

4.75.0 (2024-12-03)

Full Changelog: v4.74.0...v4.75.0

Features

• improve docs for jsr README.md (#1208) (338527e)

4.74.0 (2024-12-02)

Full Changelog: v4.73.1...v4.74.0

Features

• internal: make git install file structure match npm (#1204) (e7c4c6d)

4.73.1 (2024-11-25)

Full Changelog: v4.73.0...v4.73.1

Documentation

• readme: mention .withResponse() for streaming request ID (#1202) (b6800d4)

4.73.0 (2024-11-20)

Full Changelog: v4.72.0...v4.73.0

Features

• api: add gpt-4o-2024-11-20 model (#1201) (0feeafd)
• bump model in all example snippets to gpt-4o (6961c37)

Bug Fixes

• docs: add missing await to pagination example (#1190) (524b9e8)

Chores

• client: drop unused devDependency (#1191) (8ee6c03)
• internal: spec update (#1195) (12f9334)
• internal: use reexports not destructuring (#1181) (f555dd6)

Documentation

• bump models in example snippets to gpt-4o (#1184) (4ec4027)
• change readme title (#1198) (e34981c)
• improve jsr documentation (#1197) (ebdb4f7)
• readme: fix incorrect fileBatches.uploadAndPoll params (#1200) (3968ef1)

4.72.0 (2024-11-12)

Full Changelog: v4.71.1...v4.72.0

Features

• add back deno runtime testing without type checks (1626cf5)

Chores

• ecosystem-tests: bump wrangler version (#1178) (4dfb0c6)

4.71.1 (2024-11-06)

Full Changelog: v4.71.0...v4.71.1

Bug Fixes

• change release please configuration for jsr.json (#1174) (c39efba)

4.71.0 (2024-11-04)

Full Changelog: v4.70.3...v4.71.0

Features

• api: add support for predicted outputs (#1172) (08a7bb4)

4.70.3 (2024-11-04)

Full Changelog: v4.70.2...v4.70.3

Bug Fixes

• change streaming helper imports to be relative (e73b7cf)

4.70.2 (2024-11-01)

Full Changelog: v4.70.1...v4.70.2

Bug Fixes

• add permissions to github workflow (ee75e00)
• skip deno ecosystem test (5b181b0)

4.70.1 (2024-11-01)

Full Changelog: v4.70.0...v4.70.1

Bug Fixes

• don't require deno to run build-deno (#1167) (9d857bc)

4.70.0 (2024-11-01)

Full Changelog: v4.69.0...v4.70.0

Features

• publish to jsr (#1165) (5aa93a7)

Chores

• internal: fix isolated modules exports (9cd1958)

Refactors

• use type imports for type-only imports (#1159) (07bbaf6)

4.69.0 (2024-10-30)

Full Changelog: v4.68.4...v4.69.0

Features

• api: add new, expressive voices for Realtime and Audio in Chat Completions (#1157) (12e501c)

Bug Fixes

• internal: support pnpm git installs (#1156) (b744c5b)

Documentation

• readme: minor typo fixes (#1154) (c6c9f9a)

4.68.4 (2024-10-23)

Full Changelog: v4.68.3...v4.68.4

Chores

• internal: update spec version (#1146) (0165a8d)

4.68.3 (2024-10-23)

Full Changelog: v4.68.2...v4.68.3

Chores

• internal: bumps eslint and related dependencies (#1143) (2643f42)

4.68.2 (2024-10-22)

Full Changelog: v4.68.1...v4.68.2

Chores

• internal: update spec version (#1141) (2ccb3e3)

4.68.1 (2024-10-18)

Full Changelog: v4.68.0...v4.68.1

Bug Fixes

• client: respect x-stainless-retry-count default headers (#1138) (266717b)

4.68.0 (2024-10-17)

Full Changelog: v4.67.3...v4.68.0

Features

• api: add gpt-4o-audio-preview model for chat completions (#1135) (17a623f)

4.67.3 (2024-10-08)

Full Changelog: v4.67.2...v4.67.3

Chores

• internal: pass props through internal parser (#1125) (5ef8aa8)

4.67.2 (2024-10-07)

Full Changelog: v4.67.1...v4.67.2

Chores

• internal: move LineDecoder to a separate file (#1120) (0a4be65)

4.67.1 (2024-10-02)

Full Changelog: v4.67.0...v4.67.1

Documentation

• improve and reference contributing documentation (#1115) (7fa30b3)

4.67.0 (2024-10-01)

Full Changelog: v4.66.1...v4.67.0

Features

• api: support storing chat completions, enabling evals and model distillation in the dashboard (#1112) (6424924)

4.66.1 (2024-09-30)

Full Changelog: v4.66.0...v4.66.1

Bug Fixes

• audio: add fallback overload types (0c00a13)
• audio: use export type (1519100)

4.66.0 (2024-09-27)

Full Changelog: v4.65.0...v4.66.0

Features

• client: add request_id to .withResponse() (#1095) (2d0f565)

Bug Fixes

• audio: correct types for transcriptions / translations (#1104) (96e86c2)
• client: correct types for transcriptions / translations (#1105) (fa16ebb)

4.65.0 (2024-09-26)

Full Changelog: v4.64.0...v4.65.0

Features

• api: add omni-moderation model (#1100) (66c0f21)

4.64.0 (2024-09-25)

Full Changelog: v4.63.0...v4.64.0

Features

• client: allow overriding retry count header (#1098) (a466ff7)

Bug Fixes

• audio: correct response_format translations type (#1097) (9a5f461)

Chores

• internal: fix ecosystem tests error output (#1096) (ecdb4e9)
• internal: fix slow ecosystem test (#1093) (80ed9ec)

4.63.0 (2024-09-20)

Full Changelog: v4.62.1...v4.63.0

Features

• client: send retry count header (#1087) (7bcebc0)

Chores

• types: improve type name for embedding models (#1089) (d6966d9)

4.62.1 (2024-09-18)

Full Changelog: v4.62.0...v4.62.1

Bug Fixes

• types: remove leftover polyfill usage (#1084) (b7c9538)

4.62.0 (2024-09-17)

Full Changelog: v4.61.1...v4.62.0

Features

• client: add .requestid property to object responses (#1078) (d5c2131)

Chores

• internal: add ecosystem test for qs reproduction (0199dd8)
• internal: add query string encoder (#1079) (f870682)
• internal: fix some types (#1082) (1ec41a7)
• tests: add query string tests to ecosystem tests (36be724)

4.61.1 (2024-09-16)

Full Changelog: v4.61.0...v4.61.1

Bug Fixes

• runTools: correct request options type (#1073) (399f971)

Chores

• internal: update spec link (#1076) (20f1bcc)

4.61.0 (2024-09-13)

Full Changelog: v4.60.1...v4.61.0

Bug Fixes

• client: partial parsing update to handle strings (46e8eb6)
• examples: handle usage chunk in tool call streaming (#1068) (e4188c4)

Chores

• examples: add a small delay to tool-calls example streaming (a3fc659)

Documentation

• update CONTRIBUTING.md (#1071) (5de81c9)

4.60.1 (2024-09-13)

Full Changelog: v4.60.0...v4.60.1

Bug Fixes

• zod: correctly add $ref definitions for transformed schemas (#1065) (9b93b24)

4.60.0 (2024-09-12)

Full Changelog: v4.59.0...v4.60.0

Features

• api: add o1 models (#1061) (224cc04)

4.59.0 (2024-09-11)

Full Changelog: v4.58.2...v4.59.0

Features

• structured outputs: support accessing raw responses (#1058) (af17697)

Documentation

• azure: example for custom base URL (#1055) (20defc8)
• azure: remove locale from docs link (#1054) (f9b7eac)

4.58.2 (2024-09-09)

Full Changelog: v4.58.1...v4.58.2

Bug Fixes

• errors: pass message through to APIConnectionError (#1050) (5a34316)

Chores

• better object fallback behaviour for casting errors (#1053) (b7d4619)

4.58.1 (2024-09-06)

Full Changelog: v4.58.0...v4.58.1

Chores

• docs: update browser support information (#1045) (d326cc5)

4.58.0 (2024-09-05)

Full Changelog: v4.57.3...v4.58.0

Features

• vector store: improve chunking strategy type names (#1041) (471cec3)

Bug Fixes

• uploads: avoid making redundant memory copies (#1043) (271297b)

4.57.3 (2024-09-04)

Full Changelog: v4.57.2...v4.57.3

Bug Fixes

• helpers/zod: avoid import issue in certain environments (#1039) (e238daa)

Chores

• internal: minor bump qs version (#1037) (8ec218e)

4.57.2 (2024-09-04)

Full Changelog: v4.57.1...v4.57.2

Chores

• internal: dependency updates (#1035) (e815fb6)

4.57.1 (2024-09-03)

Full Changelog: v4.57.0...v4.57.1

Bug Fixes

• assistants: correctly accumulate tool calls when streaming (#1031) (d935ad3)
• client: correct File construction from node-fetch Responses (#1029) (22ebdc2)
• runTools without stream should not emit user message events (#1005) (22ded4d)

Chores

• internal/tests: workaround bug in recent types/node release (3c7bdfd)

4.57.0 (2024-08-29)

Full Changelog: v4.56.2...v4.57.0

Features

• api: add file search result details to run steps (#1023) (d9acd0a)

Bug Fixes

• install examples deps as part of bootstrap script (#1022) (eae8e36)

4.56.2 (2024-08-29)

Full Changelog: v4.56.1...v4.56.2

Chores

• run tsc as part of lint script (#1020) (4942347)

4.56.1 (2024-08-27)

Full Changelog: v4.56.0...v4.56.1

Chores

• ci: check for build errors (#1013) (7ff2127)

4.56.0 (2024-08-16)

Full Changelog: v4.55.9...v4.56.0

Features

• api: add chatgpt-4o-latest model (edc4398)

4.55.9 (2024-08-16)

Full Changelog: v4.55.8...v4.55.9

Bug Fixes

• azure/tts: avoid stripping model param (#999) (c3a7ccd)

4.55.8 (2024-08-15)

Full Changelog: v4.55.7...v4.55.8

Chores

• types: define FilePurpose enum (#997) (19b941b)

4.55.7 (2024-08-13)

Full Changelog: v4.55.6...v4.55.7

Bug Fixes

• json-schema: correct handling of nested recursive schemas (#992) (ac309ab)

4.55.6 (2024-08-13)

Full Changelog: v4.55.5...v4.55.6

Bug Fixes

• zod-to-json-schema: correct licensing (#986) (bd2051e)

4.55.5 (2024-08-12)

Full Changelog: v4.55.4...v4.55.5

Chores

• examples: minor formatting changes (#987) (8e6b615)
• sync openapi url (#989) (02ff1c5)

4.55.4 (2024-08-09)

Full Changelog: v4.55.3...v4.55.4

Bug Fixes

• helpers/zod: nested union schema extraction (#979) (31b05aa)

Chores

• ci: bump prism mock server version (#982) (7442643)
• ci: codeowners file (#980) (17a42b2)

4.55.3 (2024-08-08)

Full Changelog: v4.55.2...v4.55.3

Chores

• internal: updates (#975) (313a190)

4.55.2 (2024-08-08)

Full Changelog: v4.55.1...v4.55.2

Bug Fixes

• helpers/zod: add extract-to-root ref strategy (ef3c73c)
• helpers/zod: add nullableStrategy option (ad89892)
• helpers/zod: correct logic for adding root schema to definitions (e4a247a)

Chores

• internal: add README for vendored zod-to-json-schema (d8a80a9)
• tests: add more API request tests (04c1590)

4.55.1 (2024-08-07)

Full Changelog: v4.55.0...v4.55.1

Bug Fixes

• helpers/zod: correct schema generation for recursive schemas (cb54d93)

Chores

• api: remove old AssistantResponseFormat type (#967) (9fd94bf)
• internal: update test snapshots (bceea60)
• vendor/zodJsonSchema: add option to duplicate top-level ref (84b8a38)

Documentation

• examples: add UI generation example script (c75c017)

4.55.0 (2024-08-06)

Full Changelog: v4.54.0...v4.55.0

Features

• api: add structured outputs support (573787c)

4.54.0 (2024-08-02)

Full Changelog: v4.53.2...v4.54.0

Features

• extract out ImageModel, AudioModel, SpeechModel (#964) (1edf957)
• make enums not nominal (#965) (0dd0cd1)

Chores

• ci: correctly tag pre-release npm packages (#963) (f1a4a68)
• internal: add constant for default timeout (#960) (55c01f4)
• internal: cleanup event stream helpers (#950) (8f49956)

Documentation

• README: link Lifecycle in Polling Helpers section (#962) (c610c81)

4.53.2 (2024-07-26)

Full Changelog: v4.53.1...v4.53.2

Chores

• docs: fix incorrect client var names (#955) (cc91be8)

4.53.1 (2024-07-25)

Full Changelog: v4.53.0...v4.53.1

Bug Fixes

• compat: remove ReadableStream polyfill redundant since node v16 (#954) (78b2a83)

Chores

• tests: update prism version (#948) (9202c91)

4.53.0 (2024-07-22)

Full Changelog: v4.52.7...v4.53.0

Features

• api: add new gpt-4o-mini models (#942) (7ac10dd)
• api: add uploads endpoints (#946) (8709ceb)

Chores

• docs: mention support of web browser runtimes (#938) (123d19d)
• docs: use client instead of package name in Node examples (#941) (8b5db1f)

4.52.7 (2024-07-11)

Full Changelog: v4.52.6...v4.52.7

Documentation

• examples: update example values (#933) (92512ab)

4.52.6 (2024-07-11)

Full Changelog: v4.52.5...v4.52.6

Chores

• ci: also run workflows for PRs targeting next (#931) (e3f979a)

4.52.5 (2024-07-10)

Full Changelog: v4.52.4...v4.52.5

Bug Fixes

• vectorStores: correctly handle missing files in uploadAndPoll() (#926) (945fca6)

4.52.4 (2024-07-08)

Full Changelog: v4.52.3...v4.52.4

Refactors

• examples: removedduplicated 'messageDelta' streaming event. (#909) (7b0b3d2)

4.52.3 (2024-07-02)

Full Changelog: v4.52.2...v4.52.3

Chores

• minor change to tests (#916) (b8a33e3)

4.52.2 (2024-06-28)

Full Changelog: v4.52.1...v4.52.2

Chores

• gitignore test server logs (#914) (6316720)

4.52.1 (2024-06-25)

Full Changelog: v4.52.0...v4.52.1

Chores

• doc: clarify service tier default value (#908) (e4c8100)
• internal: minor reformatting (#911) (78c9377)
• internal: re-order some imports (#904) (dbd5c40)

4.52.0 (2024-06-18)

Full Changelog: v4.51.0...v4.52.0

Features

• api: add service tier argument for chat completions (#900) (91e6651)

4.51.0 (2024-06-12)

Full Changelog: v4.50.0...v4.51.0

Features

• api: updates (#894) (b58f5a1)

4.50.0 (2024-06-10)

Full Changelog: v4.49.1...v4.50.0

Features

• support application/octet-stream request bodies (#892) (51661c8)

4.49.1 (2024-06-07)

Full Changelog: v4.49.0...v4.49.1

Bug Fixes

• remove erroneous thread create argument (#889) (a9f898e)

4.49.0 (2024-06-06)

Full Changelog: v4.48.3...v4.49.0

Features

• api: updates (#887) (359eeb3)

4.48.3 (2024-06-06)

Full Changelog: v4.48.2...v4.48.3

Chores

• internal: minor refactor of tests (#884) (0b71f2b)

4.48.2 (2024-06-05)

Full Changelog: v4.48.1...v4.48.2

Chores

• internal: minor change to tests (#881) (5e2d608)

4.48.1 (2024-06-04)

Full Changelog: v4.48.0...v4.48.1

Bug Fixes

• resolve typescript issue (1129707)

4.48.0 (2024-06-03)

Full Changelog: v4.47.3...v4.48.0

Features

• api: updates (#874) (295c248)

4.47.3 (2024-05-31)

Full Changelog: v4.47.2...v4.47.3

Bug Fixes

• allow git imports for pnpm (#873) (9da9809)

Documentation

• azure: update example and readme to use Entra ID (#857) (722eff1)

4.47.2 (2024-05-28)

Full Changelog: v4.47.1...v4.47.2

Documentation

• readme: add bundle size badge (#869) (e252132)

4.47.1 (2024-05-14)

Full Changelog: v4.47.0...v4.47.1

Chores

• internal: add slightly better logging to scripts (#848) (139e690)

4.47.0 (2024-05-14)

Full Changelog: v4.46.1...v4.47.0

Features

• api: add incomplete state (#846) (5f663a1)

4.46.1 (2024-05-13)

Full Changelog: v4.46.0...v4.46.1

Refactors

• change import paths to be relative (#843) (7913574)

4.46.0 (2024-05-13)

Full Changelog: v4.45.0...v4.46.0

Features

• api: add gpt-4o model (#841) (c818ed1)

4.45.0 (2024-05-11)

Full Changelog: v4.44.0...v4.45.0

Features

• azure: batch api (#839) (e279f8c)

Chores

• dependency: bumped Next.js version (#836) (babb140)
• docs: add SECURITY.md (#838) (6e556d9)

4.44.0 (2024-05-09)

Full Changelog: v4.43.0...v4.44.0

Features

• api: add message image content (#834) (7757b3e)

4.43.0 (2024-05-08)

Full Changelog: v4.42.0...v4.43.0

Features

• api: adding file purposes (#831) (a62b877)

4.42.0 (2024-05-06)

Full Changelog: v4.41.1...v4.42.0

Features

• api: add usage metadata when streaming (#829) (6707f11)

Bug Fixes

• example: fix fine tuning example (#827) (6480a50)

4.41.1 (2024-05-06)

Full Changelog: v4.41.0...v4.41.1

Bug Fixes

• azure: update build script (#825) (8afc6e7)

4.41.0 (2024-05-05)

Full Changelog: v4.40.2...v4.41.0

Features

• client: add Azure client (#822) (92f9049)

4.40.2 (2024-05-03)

Full Changelog: v4.40.1...v4.40.2

Bug Fixes

• package: revert recent client file change (#819) (fa722c9)
• vectorStores: correct uploadAndPoll method (#817) (d63f22c)

4.40.1 (2024-05-02)

Full Changelog: v4.40.0...v4.40.1

Chores

• internal: bump prism version (#813) (81a6c28)
• internal: move client class to separate file (#815) (d0b915a)

4.40.0 (2024-05-01)

Full Changelog: v4.39.1...v4.40.0

Features

• api: delete messages (#811) (9e37dbd)

4.39.1 (2024-04-30)

Full Changelog: v4.39.0...v4.39.1

Chores

• internal: add link to openapi spec (#810) (61b5b83)
• internal: fix release please for deno (#808) (ecc2eae)
• internal: refactor scripts (#806) (9283519)

4.39.0 (2024-04-29)

Full Changelog: v4.38.5...v4.39.0

Features

• api: add required tool_choice (#803) (99693e6)

Chores

• internal: add scripts/test and scripts/mock (#801) (6656105)

4.38.5 (2024-04-24)

Full Changelog: v4.38.4...v4.38.5

Chores

• internal: use actions/checkout@v4 for codeflow (#799) (5ab7780)

4.38.4 (2024-04-24)

Full Changelog: v4.38.3...v4.38.4

Bug Fixes

• api: change timestamps to unix integers (#798) (7271a6c)
• docs: doc improvements (#796) (49fcc86)

4.38.3 (2024-04-22)

Full Changelog: v4.38.2...v4.38.3

Chores

• internal: use @swc/jest for running tests (#793) (8947f19)

4.38.2 (2024-04-19)

Full Changelog: v4.38.1...v4.38.2

Bug Fixes

• api: correct types for message attachment tools (#787) (8626884)

4.38.1 (2024-04-18)

Full Changelog: v4.38.0...v4.38.1

Bug Fixes

• api: correct types for attachments (#783) (6893631)

4.38.0 (2024-04-18)

Full Changelog: v4.37.1...v4.38.0

Features

• api: batch list endpoint (#781) (d226759)

4.37.1 (2024-04-17)

Full Changelog: v4.37.0...v4.37.1

Chores

• api: docs and response_format response property (#778) (78f5c35)

4.37.0 (2024-04-17)

Full Changelog: v4.36.0...v4.37.0

Features

• api: add vector stores (#776) (8bb929b)

4.36.0 (2024-04-16)

Full Changelog: v4.35.0...v4.36.0

Features

• client: add header OpenAI-Project (#772) (bb4df37)
• extract chat models to a named enum (#775) (141d2ed)

Build System

• configure UTF-8 locale in devcontainer (#774) (bebf4f0)

4.35.0 (2024-04-15)

Full Changelog: v4.34.0...v4.35.0

Features

• errors: add request_id property (#769) (43aa6a1)

4.34.0 (2024-04-15)

Full Changelog: v4.33.1...v4.34.0

Features

• api: add batch API (#768) (7fe34f2)
• api: updates (#766) (52bcc47)

4.33.1 (2024-04-12)

Full Changelog: v4.33.0...v4.33.1

Chores

• internal: formatting (#763) (b6acf54)
• internal: improve ecosystem tests (#761) (fcf748d)

4.33.0 (2024-04-05)

Full Changelog: v4.32.2...v4.33.0

Features

• api: add additional messages when creating thread run (#759) (f1fdb41)

4.32.2 (2024-04-04)

Full Changelog: v4.32.1...v4.32.2

Bug Fixes

• streaming: handle special line characters and fix multi-byte character decoding (#757) (8dcdda2)
• tests: update wrangler to v3.19.0 (CVE-2023-7080) (#755) (47ca41d)

Chores

• tests: bump ecosystem tests dependencies (#753) (3f86ea2)

4.32.1 (2024-04-02)

Full Changelog: v4.32.0...v4.32.1

Chores

• deps: bump yarn to v1.22.22 (#751) (5b41d10)

4.32.0 (2024-04-01)

Full Changelog: v4.31.0...v4.32.0

Features

• api: add support for filtering messages by run_id (#747) (9a397ac)
• api: run polling helpers (#749) (02920ae)

Chores

• deps: remove unused dependency digest-fetch (#748) (5376837)

Documentation

• readme: change undocumented params wording (#744) (8796691)

Refactors

• rename createAndStream to stream (02920ae)

4.31.0 (2024-03-30)

Full Changelog: v4.30.0...v4.31.0

Features

• api: adding temperature parameter (#742) (b173b05)

Bug Fixes

• streaming: trigger all event handlers with fromReadableStream (#741) (7b1e593)

4.30.0 (2024-03-28)

Full Changelog: v4.29.2...v4.30.0

Features

• assistant fromReadableStream (#738) (8f4ba18)

Bug Fixes

• client: correctly send deno version header (#736) (b7ea175)
• example: correcting example (#739) (a819551)
• handle process.env being undefined in debug func (#733) (2baa149)
• internal: make toFile use input file's options (#727) (15880d7)

Chores

• internal: add type (#737) (18c1989)

Documentation

• readme: consistent use of sentence case in headings (#729) (7e515fd)
• readme: document how to make undocumented requests (#730) (a06d861)

4.29.2 (2024-03-19)

Full Changelog: v4.29.1...v4.29.2

Chores

• internal: update generated pragma comment (#724) (139e205)

Documentation

• assistant improvements (#725) (6a2c41b)
• fix typo in CONTRIBUTING.md (#722) (05ff8f7)

4.29.1 (2024-03-15)

Full Changelog: v4.29.0...v4.29.1

Documentation

• readme: assistant streaming (#719) (bc9a1ca)

4.29.0 (2024-03-13)

Full Changelog: v4.28.5...v4.29.0

Features

• assistants: add support for streaming (#714) (7d27d28)

4.28.5 (2024-03-13)

Full Changelog: v4.28.4...v4.28.5

Bug Fixes

• ChatCompletionStream: abort on async iterator break and handle errors (#699) (ac417a2)
• streaming: correctly handle trailing new lines in byte chunks (#708) (4753be2)

Chores

• api: update docs (#703) (e1db98b)
• docs: mention install from git repo (#700) (c081bdb)
• fix error handler in readme (#704) (4ff790a)
• internal: add explicit type annotation to decoder (#712) (d728e99)
• types: fix accidental exposure of Buffer type to cloudflare (#709) (0323ecb)

Documentation

• contributing: improve wording (#696) (940d569)
• readme: fix https proxy example (#705) (d144789)
• readme: fix typo in custom fetch implementation (#698) (64041fd)
• remove extraneous --save and yarn install instructions (#710) (8ec216d)
• use @deprecated decorator for deprecated params (#711) (4688ef4)

4.28.4 (2024-02-28)

Full Changelog: v4.28.3...v4.28.4

Features

• api: add wav and pcm to response_format (#691) (b1c6171)

Chores

• ci: update actions/setup-node action to v4 (#685) (f2704d5)
• internal: fix ecosystem tests (#693) (616624d)
• types: extract run status to a named type (#686) (b3b3b8e)
• update @types/react to 18.2.58, @types/react-dom to 18.2.19 (#688) (2a0d0b1)
• update dependency @types/node to v20.11.20 (#690) (4ca005b)
• update dependency @types/ws to v8.5.10 (#683) (a617268)
• update dependency next to v13.5.6 (#689) (abb3b66)

4.28.3 (2024-02-20)

Full Changelog: v4.28.2...v4.28.3

Bug Fixes

• ci: revert "move github release logic to github app" (#680) (8b4009a)

4.28.2 (2024-02-19)

Full Changelog: v4.28.1...v4.28.2

Bug Fixes

• api: remove non-GA instance_id param (#677) (4d0d4da)

4.28.1 (2024-02-19)

Full Changelog: v4.28.0...v4.28.1

Chores

• ci: move github release logic to github app (#671) (ecca6bc)
• internal: refactor release environment script (#674) (27d3770)

4.28.0 (2024-02-13)

Full Changelog: v4.27.1...v4.28.0

Features

• api: updates (#669) (e1900f9)

4.27.1 (2024-02-12)

Full Changelog: v4.27.0...v4.27.1

4.27.0 (2024-02-08)

Full Changelog: v4.26.1...v4.27.0

Features

• api: add timestamp_granularities, add gpt-3.5-turbo-0125 model (#661) (5016806)

Chores

• internal: fix retry mechanism for ecosystem-test (#663) (0eb7ed5)
• respect application/vnd.api+json content-type header (#664) (f4fad54)

4.26.1 (2024-02-05)

Full Changelog: v4.26.0...v4.26.1

Chores

• internal: enable building when git installed (#657) (8c80a7d)
• internal: re-order pagination import (#656) (21ae54e)
• internal: support pre-release versioning (#653) (0c3859f)
• test: add delay between ecosystem tests retry (#651) (6a4cc5c)

Documentation

• add a CONTRIBUTING.md (#659) (8ea58b0)

4.26.0 (2024-01-25)

Full Changelog: v4.25.0...v4.26.0

Features

• api: add text embeddings dimensions param (#650) (1b5a977)

Chores

• internal: add internal helpers & improve build scripts (#643) (9392f50)
• internal: adjust ecosystem-tests logging in CI (#646) (156084b)
• internal: don't re-export streaming type (#648) (4c4be94)
• internal: fix binary files (#645) (e1fbc39)
• internal: minor streaming updates (#647) (2f073e4)
• internal: pin deno version (#649) (7e4b903)

4.25.0 (2024-01-21)

Full Changelog: v4.24.7...v4.25.0

Features

• api: add usage to runs and run steps (#640) (3caa416)

Bug Fixes

• allow body type in RequestOptions to be null (#637) (c4f8a36)
• handle system_fingerprint in streaming helpers (#636) (f273530)
• types: accept undefined for optional client options (#635) (e48cd57)

Chores

• internal: debug logging for retries; speculative retry-after-ms support (#633) (fd64971)
• internal: update comment (#631) (e109d40)

4.24.7 (2024-01-13)

Full Changelog: v4.24.6...v4.24.7

Chores

• ecosystem-tests: fix flaky vercel-edge, cloudflare-worker, and deno tests (#626) (ae412a5)
• ecosystem-tests: fix typo in deno test (#628) (048ec94)

4.24.6 (2024-01-12)

Full Changelog: v4.24.5...v4.24.6

Chores

• ecosystem-tests: fix flaky tests and remove fine tuning calls (#623) (258d79f)
• ecosystem-tests: fix flaky tests and remove fine tuning calls (#625) (58e5fd8)

4.24.5 (2024-01-12)

Full Changelog: v4.24.4...v4.24.5

Refactors

• api: remove deprecated endpoints (#621) (2054d71)

4.24.4 (2024-01-11)

Full Changelog: v4.24.3...v4.24.4

Chores

• internal: narrow type into stringifyQuery (#619) (88fb9cd)

4.24.3 (2024-01-10)

Full Changelog: v4.24.2...v4.24.3

Bug Fixes

• use default base url if BASE_URL env var is blank (#615) (a27ad3d)

4.24.2 (2024-01-08)

Full Changelog: v4.24.1...v4.24.2

Bug Fixes

• headers: always send lowercase headers and strip undefined (BREAKING in rare cases) (#608) (4ea159f)

Chores

• add .keep files for examples and custom code directories (#612) (5e0f733)
• internal: bump license (#605) (045ee74)
• internal: improve type signatures (#609) (e1ccc82)

Documentation

• fix docstring typos (#600) (1934fa1)
• improve audio example to show how to stream to a file (#598) (e950ad9)

4.24.1 (2023-12-22)

Full Changelog: v4.24.0...v4.24.1

Bug Fixes

• pagination: correct type annotation object field (#590) (4066eda)

Documentation

• messages: improvements to helpers reference + typos (#595) (96a59b9)
• reformat README.md (#592) (8ffc7f8)

Refactors

• write jest config in typescript (#588) (eb6ceeb)

4.24.0 (2023-12-19)

Full Changelog: v4.23.0...v4.24.0

Features

• api: add additional instructions for runs (#586) (401d93e)

Chores

• deps: update dependency start-server-and-test to v2.0.3 (#580) (8e1aca1)
• deps: update dependency ts-jest to v29.1.1 (#578) (a6edb7b)
• deps: update jest (#582) (e49e471)
• internal: bump deps (#583) (2e07b4c)
• internal: update deps (#581) (7b690dc)

Documentation

• upgrade models in examples to latest version (#585) (60101a4)

4.23.0 (2023-12-17)

Full Changelog: v4.22.1...v4.23.0

Features

• api: add token logprobs to chat completions (#576) (8d4292e)

Chores

• ci: run release workflow once per day (#574) (529f09f)

4.22.1 (2023-12-15)

Full Changelog: v4.22.0...v4.22.1

Chores

• update dependencies (#572) (a51e620)

Documentation

• replace runFunctions with runTools in readme (#570) (c3b9ad5)

4.22.0 (2023-12-15)

Full Changelog: v4.21.0...v4.22.0

Features

• api: add optional name argument + improve docs (#569) (3b68ace)

Chores

• update prettier (#567) (83dec2a)

4.21.0 (2023-12-11)

Full Changelog: v4.20.1...v4.21.0

Features

• client: support reading the base url from an env variable (#547) (06fb68d)

Bug Fixes

• correct some runTools behavior and deprecate runFunctions (#562) (f5cdd0f)
• prevent 400 when using runTools/runFunctions with Azure OpenAI API (#544) (735d9b8)

Documentation

• readme: update example snippets (#546) (566d290)

Build System

• specify packageManager: yarn (#561) (935b898)

4.20.1 (2023-11-24)

Full Changelog: v4.20.0...v4.20.1

Chores

• internal: remove file import and conditionally run prepare (#533) (48cb729)

Documentation

• readme: fix typo and add examples link (#529) (cf959b1)

4.20.0 (2023-11-22)

Full Changelog: v4.19.1...v4.20.0

Features

• allow installing package directly from github (#522) (51926d7)

Chores

• internal: don't call prepare in dist (#525) (d09411e)

4.19.1 (2023-11-20)

Full Changelog: v4.19.0...v4.19.1

4.19.0 (2023-11-15)

Full Changelog: v4.18.0...v4.19.0

Features

• api: updates (#501) (944d58e)

4.18.0 (2023-11-14)

Full Changelog: v4.17.5...v4.18.0

Features

• api: add gpt-3.5-turbo-1106 (#496) (45f7672)

4.17.5 (2023-11-13)

Full Changelog: v4.17.4...v4.17.5

Chores

• fix typo in docs and add request header for function calls (#494) (22ce244)

4.17.4 (2023-11-10)

Full Changelog: v4.17.3...v4.17.4

Chores

• internal: update jest config (#482) (3013e8c)

4.17.3 (2023-11-09)

Full Changelog: v4.17.2...v4.17.3

4.17.2 (2023-11-09)

Full Changelog: v4.17.1...v4.17.2

Chores

• internal: bump deno version number (#478) (69913f3)

4.17.1 (2023-11-09)

Full Changelog: v4.17.0...v4.17.1

Refactors

• client: deprecate files.retrieveContent in favour of files.content (#474) (7c7bfc2)

4.17.0 (2023-11-08)

Full Changelog: v4.16.2...v4.17.0

Features

• api: unify function types (#467) (d51cd94)

Refactors

• api: rename FunctionObject to FunctionDefinition (#470) (f3990c7)

4.16.2 (2023-11-08)

Full Changelog: v4.16.1...v4.16.2

Bug Fixes

• api: accidentally required params, add new models & other fixes (#463) (1cb403e)
• api: update embedding response object type (#466) (53b7e25)
• asssitant_deleted -> assistant_deleted (#452) (ef89bd7)
• types: ensure all code paths return a value (#458) (19402c3)

Chores

• docs: fix github links (#457) (6b9b94e)
• internal: fix typo in comment (#456) (fe24342)

Documentation

• update deno deploy link to include v (#441) (47b13aa)

4.16.1 (2023-11-06)

Full Changelog: v4.16.0...v4.16.1

Bug Fixes

• api: retreival -> retrieval (#437) (b4bd3ee)

Documentation

• api: improve docstrings (#435) (ee8b24c)

4.16.0 (2023-11-06)

Full Changelog: v4.15.4...v4.16.0

Features

• api: releases from DevDay; assistants, multimodality, tools, dall-e-3, tts, and more (#433) (fb92f5e)

Bug Fixes

• improve deno readme (#429) (871ceac)

Documentation

• deno version (#432) (74bf336)
• update deno link in more places (#431) (5da63d4)

4.15.4 (2023-11-05)

Full Changelog: v4.15.3...v4.15.4

Documentation

• readme: remove redundant whitespace (#427) (aa3a178)

4.15.3 (2023-11-04)

Full Changelog: v4.15.2...v4.15.3

Bug Fixes

• improve deno releases (#425) (19469f2)

4.15.2 (2023-11-04)

Full Changelog: v4.15.1...v4.15.2

Documentation

• fix deno.land import (#423) (e5415a2)

4.15.1 (2023-11-04)

Full Changelog: v4.15.0...v4.15.1

Documentation

• document customizing fetch (#420) (1ca982f)

4.15.0 (2023-11-03)

Full Changelog: v4.14.2...v4.15.0

Features

• beta: add streaming and function calling helpers (#409) (510c1f3)
• client: allow binary returns (#416) (02f7ad7)
• github: include a devcontainer setup (#413) (fb2996f)
• streaming improvements (#411) (37b622c)

4.14.2 (2023-10-30)

Full Changelog: v4.14.1...v4.14.2

Chores

• docs: update deno link (#407) (0328882)

4.14.1 (2023-10-27)

Full Changelog: v4.14.0...v4.14.1

Bug Fixes

• deploy deno in a github workflow instead of postpublish step (#405) (3a6dba0)
• typo in build script (#403) (76c5c96)

Chores

• internal: update gitignore (#406) (986b0bb)

4.14.0 (2023-10-25)

Full Changelog: v4.13.0...v4.14.0

Features

• client: adjust retry behavior to be exponential backoff (#400) (2bc14ce)

Chores

• docs: update deno version (#399) (cdee077)

4.13.0 (2023-10-22)

Full Changelog: v4.12.4...v4.13.0

Features

• api: add embeddings encoding_format (#390) (cf70dea)
• handle 204 No Content gracefully (#391) (2dd005c)

4.12.4 (2023-10-17)

Full Changelog: v4.12.3...v4.12.4

Bug Fixes

• import web-streams-polyfill without overriding globals (#385) (be8e18b)

4.12.3 (2023-10-16)

Full Changelog: v4.12.2...v4.12.3

Documentation

• organisation -> organization (UK to US English) (#382) (516f0ad)

4.12.2 (2023-10-16)

Full Changelog: v4.12.1...v4.12.2

Bug Fixes

• client: correctly handle errors during streaming (#377) (09233b1)
• client: correctly handle errors during streaming (#379) (9ced580)
• improve status code in error messages (#381) (68dfb17)

Chores

• add case insensitive get header function (#373) (b088998)
• internal: add debug logs for stream responses (#380) (689db0b)
• show deprecation notice on re-export (#368) (b176703)
• update comment (#376) (a06c685)
• update comment (#378) (b04031d)

Refactors

• streaming: change Stream constructor signature (#370) (71984ed)
• test: refactor authentication tests (#371) (e0d459f)

4.12.1 (2023-10-11)

Full Changelog: v4.12.0...v4.12.1

Bug Fixes

• fix namespace exports regression (#366) (b2b1d85)

4.12.0 (2023-10-11)

Full Changelog: v4.11.1...v4.12.0

Features

• api: remove content_filter stop_reason and update documentation (#352) (a4b401e)
• re-export chat completion types at the top level, and work around webpack limitations (#365) (bb815d0)

Bug Fixes

• prevent ReferenceError, update compatibility to ES2020 and Node 18+ (#356) (fc71a4b)

Chores

• internal: minor formatting improvement (#354) (3799863)

4.11.1 (2023-10-03)

Full Changelog: v4.11.0...v4.11.1

4.11.0 (2023-09-29)

Full Changelog: v4.10.0...v4.11.0

Features

• client: handle retry-after with a date (#340) (b6dd384)
• package: export a root error type (#338) (462bcda)

Bug Fixes

• api: add content_filter to chat completion finish reason (#344) (f10c757)

Chores

• internal: bump lock file (#334) (fd2337b)
• internal: update lock file (#339) (1bf84b6)
• internal: update lock file (#342) (0001f06)
• internal: update lock file (#343) (a02ac8e)

4.10.0 (2023-09-21)

Full Changelog: v4.9.1...v4.10.0

Features

• api: add 'gpt-3.5-turbo-instruct', fine-tune error objects, update documentation (#329) (e5f3852)

4.10.0 (2023-09-21)

Full Changelog: v4.9.1...v4.10.0

Features

• api: add 'gpt-3.5-turbo-instruct', fine-tune error objects, update documentation (#329) (e5f3852)

4.9.1 (2023-09-21)

Full Changelog: v4.9.0...v4.9.1

Documentation

• README: fix variable names in some examples (#327) (5e05b31)

4.9.0 (2023-09-20)

Full Changelog: v4.8.0...v4.9.0

Features

• client: support importing node or web shims manually (#325) (628f293)

4.8.0 (2023-09-15)

Full Changelog: v4.7.1...v4.8.0

Features

• errors: add status code to error message (#315) (9341219)

4.7.1 (2023-09-15)

Full Changelog: v4.7.0...v4.7.1

Documentation

• declare Bun 1.0 officially supported (#314) (a16e268)

4.7.0 (2023-09-14)

Full Changelog: v4.6.0...v4.7.0

Features

• client: retry on 408 Request Timeout (#310) (1f98eac)
• make docs urls in comments absolute (#306) (9db3819)

4.6.0 (2023-09-08)

Full Changelog: v4.5.0...v4.6.0

Features

• types: extract ChatCompletionRole enum to its own type (#298) (5893e37)

Bug Fixes

• fix module not found errors in Vercel edge (#300) (47c79fe)

4.5.0 (2023-09-06)

Full Changelog: v4.4.0...v4.5.0

Features

• client: add files.waitForProcessing() method (#292) (ef59010)
• fixes tests where an array has to have unique enum values (#290) (a10b895)
• make docs more readable by eliminating unnecessary escape sequences (#287) (a068043)

Bug Fixes

• client: fix TS errors that appear when users Go to Source in VSCode (#281) (8dc59bc), closes #249
• client: handle case where the client is instantiated with a undefined baseURL (#285) (5095cf3)
• client: use explicit file extensions in _shims imports (#276) (16fe929)

Documentation

• api: update docstrings (#286) (664e953)
• readme: add link to api.md (#291) (0d1cce2)

4.4.0 (2023-09-01)

Full Changelog: v4.3.1...v4.4.0

Features

• package: add Bun export map (#269) (16f239c)
• re-export chat completion types at the top level (#268) (1a71a39)
• tests: unskip multipart form data tests (#275) (47d3e18)
• types: fix ambiguous auto-import for chat completions params (#266) (19c99fb)

Bug Fixes

• revert import change which triggered circular import bug in webpack (#274) (6534e36)

4.3.1 (2023-08-29)

Full Changelog: v4.3.0...v4.3.1

Bug Fixes

• types: improve getNextPage() return type (#262) (245a984)

Chores

• ci: setup workflows to create releases and release PRs (#259) (290908c)

4.3.0 (2023-08-27)

Features

• client: add auto-pagination to fine tuning list endpoints (#254) (5f89c5e)
• cli: rewrite in JS for better compatibility (#244) (d8d7c05)

Bug Fixes

• stream: declare Stream.controller as public (#252) (81e5de7)

Documentation

• readme: mention Azure support (#253) (294727a)

Chores

• internal: add helper method (#255) (6d8cff0)

4.2.0 (2023-08-23)

Features

• types: export RequestOptions type (#240) (ecf3bce)

Chores

• internal: export HeadersInit type shim (#241) (cf9f672)