NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
npm
npm, #1698 most downloaded on npm
The official TypeScript library for the OpenAI API
Latest stable
7.55.0
Last release 12 days ago
10 Nov 2026
Ships on a steady schedule
a new release about every 9 days
Nearly every release is documented
notes for 60 of the last 60 stable releases
Nothing withdrawn
no release was ever pulled
6 years old
426 releases · first in 2020
Drag across the ruler to compare, or tap a release to open it.
api: add typed async function tools
Your coding agent can read these notes before it upgrades. Set up the MCP server →
api: recognize gpt-6-rosalind in beta Responses
agents: deliver session errors as stream events
api: add Australia, India, and Japan residency options
api: add 12-hour Responses caching and cache-write usage
api: add workload identity providers and federation rules
api: expand Live Responses delegation tools and choices (exhaustive TypeScript switches may need to handle new variants)
api: support typed Responses WebSocket warmup
api: add agent context compaction settings
api: add gpt-6.1-terra model identifier
api: expand Live Decision item and question types. Breaking change in a minor release. Live Decision item events now include messages and function/cus…
client.beta.agents.automations, including invocation endpoints and the automations agent-tool variant (#3029) (d8931e7)DecisionItemCreatedEvent.item and DecisionItemCreateEvent.item are unions, so callers that directly access item.content or item.role may no longer compile. For both event types, use if ('content' in event.item) before accessing message fields, and handle tool-item variants separately. Message items may omit type, so checking only item.type === 'message' can skip valid messages. This release also updates image-model documentation: gpt-image-1 was shut down on October 23, 2026; replace explicit selections with gpt-image-2.5-sunburst or gpt-image-2.5-flare. The documented default for Responses image-generation tools is now gpt-image-2; specify a supported model explicitly when you need a particular model. (#3034) (7562cd2)api: Breaking change in a minor release. A Live response's Decisions question instructions can now be a string or an array of text, image, or JSON par…
instructions can now be a string or an array of text, image, or JSON parts. Code such as question.instructions.trim() can fail when it receives an array; check typeof question.instructions === 'string' and handle the array case. Decisions request inputs also accept new message roles and tool-history items, so update code that assumes it knows every input type. DecisionInputMessage is now a union: code that extends or implements the old interface must use DecisionInputMessage.DecisionInputUserMessage instead. Existing string instructions and user-message inputs still work. (21a056d)add Vertex AI support for the Responses API
api: add 30-minute cache-write usage counts
api: add typed Live sideband events
api: add degraded external storage status
api: add safety identifiers to image requests and responses
api: add Decisions prompt caching options
api: add regional processing residency options
add browser and computer toolsets (1/4)
api: add agent environment suspension and expiration
api: add prewarmed hosted environments
api: correct custom voice creation parameters
api: add standalone Decisions support
api: add custom voice creation and agent session events
agents: prepare hosted files and download result artifacts
collect final output from beta Agents streams
api: add computer use to beta agents
api: add Agents credential and session options
api: add GPT-6 Sol and Luna model identifiers
api: add session environment reset events
api: add environment-variable vault credentials
api: add agent session model settings
api: add compaction progress events
websocket: add per-iterator incoming event limits
## 7.15.0 (2026-09-10) ### Features * api: add Agents API
deps-dev: bump joi from 18.2.3 to 18.2.5 in /ecosystem-tests/vercel-edge
api: Add API key expiration controls
Includes the GPT Image 2.5 support from 7.12.0, which was not published to npm.
api: update usage APIs and documentation
api: add compute_units to Responses and Chat Completions usage
api: Add obfuscation field to ChatCompletionChunk
deps: remediate Dependabot and fixture vulnerabilities
api: Add gpt-5.5 model and tool metadata fields
organize SDK guides and repository metadata
Full Changelog: v7.2.0...v7.3.0
## 7.2.0 (2026-07-30) Full Changelog: v7.1.0...v7.2.0 ### Features * api: fast tier
api: transcription model updates
Full Changelog: v7.0.0...v7.1.0
require Node.js 22 and codify version support
Full Changelog: v6.49.0...v7.0.0
api: accept None for prompt_cache_key/safety_identifier
Full Changelog: v6.48.0...v6.49.0
None for prompt_cache_key/safety_identifier (c54a7f3)spend_limit admin apis (e8238b0)api: /organization/projects/{project_id}/service_accounts/{service_account_id}/api_keys" endpoint
Full Changelog: v6.47.0...v6.48.0
add fromReadableStream to ResponseStream
Full Changelog: v6.46.0...v6.47.0
assistants: place array delta entries by index instead of appending
Full Changelog: v6.45.0...v6.46.0
One column per quarter.