OpenAI Python SDK v3.13.0 Adds Support for New Agents API
OpenAI shipped v3.13.0 of its openai-python library, adding support for a new Agents API. The release is a client-library update, not a new model.
OpenAI Python SDK v3.13.0 Adds Agents API Support
OpenAI released version 3.13.0 of its official openai-python library on September 10, per the GitHub release tag. The update's sole listed feature is client support for a new Agents API (commit 1c4284a).
What changed
The release notes are minimal, consistent with prior patch-style updates to the SDK:
- api: add Agents API — the Python client now exposes methods for the Agents API endpoint
No further detail is provided in the changelog about the Agents API's scope, request/response schema, supported models, or pricing. OpenAI has not published accompanying documentation in this release, so the practical capabilities of the endpoint remain according to OpenAI's commit message alone rather than a full specification.
Context
This is a library release, not a model release. The openai-python repository ships incremental updates to the official Python client that wraps OpenAI's REST API — version bumps here typically track new endpoints, bug fixes, or type updates rather than new model weights. Nothing in this release indicates a new foundation model, updated context window, or revised pricing for any OpenAI model.
The versioning jump follows OpenAI's pattern of frequent minor SDK releases (v3.x series), reflecting ongoing expansion of the API surface rather than a single monolithic model update.
What this means
Developers using openai-python can now call an Agents API directly through the official client, once OpenAI publishes matching documentation for the endpoint. Until OpenAI details the Agents API's behavior, supported models, and pricing, teams should treat this as an early access point rather than a production-ready feature. The absence of specifics in the release notes — no context window, no benchmark, no pricing per 1M tokens — means this update is best understood as plumbing: it enables future agent-building workflows in code, but the substance of what an "agent" call actually does on OpenAI's backend has not yet been disclosed.
Related Articles
OpenAI Python SDK 3.35.0 adds ultrafast service tier for Agents and typed Live sideband events
OpenAI released version 3.35.0 of its Python SDK on October 17, 2026. It adds an ultrafast service tier for Agents, typed Live sideband events, and a realtime bug fix for websockets 13. Pricing and tier specifications were not disclosed in the release notes.
OpenAI red-team agents built a CPU cache covert channel across microVM sandboxes with no network or shared files
According to an OpenAI blog post, agents in separate microVMs on shared hosts exchanged synthetic secrets by manipulating the CPU cache, with no network link or shared writable files. Three other channels also turned up, and the tests used an upcoming GPT-6 Astra Cyber model and an internal frontier model.
OpenAI Paused Its Most Capable Training Workloads for 11 Days After September 20 Containment Incident
OpenAI published an account on October 23, 2026 of how it hardened the environments used to train and evaluate frontier models. It disclosed a standing no-internet training policy, an 11-day pause of its most capable training workloads after a September 20, 2026 containment incident, and a red-teaming campaign using fleets of frontier-model agents.
OpenAI upgrades ChatGPT Chrome extension with multi-tab awareness and ChatGPT Work site control
OpenAI has upgraded its ChatGPT Chrome extension with a sidebar that reads across open tabs. Through ChatGPT Work, it can also control Chrome tabs to send messages, fill out forms and download files. TechRadar tested the form-filling feature and found it useful but in need of supervision.
Comments
Loading...