Back to Explore

openai/openai-python

GitHub
60 updates · last 90 daysOpen source

Last release:

OpenAI’s official Python library for building applications that use the OpenAI API. Useful for developers who want to authenticate and make API requests from Python.

Project status

  • Actively maintained, with an upstream push today and several updates in the past week.
  • The apparent cadence is very frequent, often multiple updates in a day.

AI summary generated

AI-generated from public sources. May be inaccurate. Report

Recent updates

  • v3.47.0

    Version 3.47.0 adds a model option for the Responses API. The code updates the model allowlists used by Responses create and compact operations in the standard and beta interfaces.

    Features
  • v3.46.0

    v3.46.0 adds data-residency regions for Australia, India, and Japan, catalogs GPT-6 Rosalind, and expands image-generation controls and agent-turn error reporting. The diff also adds image context-compression options and response metadata, plus SDK support for additional GPT Image models and SVG usage reporting.

    Features
  • v3.45.0

    Version 3.45.0 adds longer prompt-cache retention and cache-write usage details, along with project-level service-tier settings APIs. The release also clarifies ownership of generated files. No breaking changes or dependency updates are evident in the supplied diff.

    Features
  • v3.44.0

    v3.44.0 adds admin API support for registering and managing OIDC workload identity providers and their federation rules. The code also expands ChatKit workflow state variables to support integer values, a change not mentioned in the release notes.

    Features
  • v3.43.0

    Version 3.43.0 expands Live Responses delegation tool support and improves WebSocket response progress accumulation and lane overflow isolation. The code adds generated types for the expanded delegation options, richer accumulator projections, and an opt-in session setting for isolating overflowing lanes.

    Features
  • v3.42.1

    Version 3.42.1 fixes beta Realtime WebSocket authentication handling and refactors stable and beta Responses WebSocket policy into shared SDK helpers. The main compatibility concern is that some provider-backed beta Realtime connections now require caller-supplied Authorization headers, while X.509 workload identity is rejected.

    Breaking
  • v3.42.0

    Version 3.42.0 adds agent configuration controls, organization external-key management, image and realtime API capabilities, and response-format updates with local output parsing support. It also improves WebSocket behavior and fixes preservation of refusal output and authentication in WebSocket helpers.

    Features
  • v3.41.0

    Version 3.41.0 adds Realtime input support for typed references, response tool usage models, metadata filtering for agent sessions, and recovery of queued binary frames after WebSocket reconnects. It also updates examples and API documentation, including notices about upcoming model and endpoint shutdowns. No immediate breaking change to the public SDK API is evident in the supplied diff.

    Features
  • v3.40.0

    Version 3.40.0 adds prompt-cache breakpoint TTL options and fixes the browser demo dependency declaration. The code diff also changes WebSocket query-parameter fallback behavior and makes internal CI workflow changes that are not clearly called out in the release notes.

    Features
  • v3.39.0

    Version 3.39.0 adds the `gpt-6.1-terra` model identifier to the Python client's response and chat model types. The diff contains no runtime behavior changes beyond accepting this identifier in the relevant typed API interfaces.

    Features
  • v3.38.0

    v3.38.0 adds API support for delegation controls, prompt cache lifetimes, and expanded Live Decision item and question types. It also fixes several parsing, tool-namespace, agent schema-warning, and API error-formatting issues. The visible diff additionally reveals changes to module-level client credential detection and to error text for non-JSON responses that are not specifically called out in the release notes.

    Features
  • v3.37.0

    v3.37.0 expands the Decisions API types, including structured question instructions and additional request input items. It also refactors upload and vector-store polling helpers and adds a WebSocket query-parameter fallback, while preserving existing helper behavior and Live connection defaults.

    BreakingFeatures
  • v3.36.1

    v3.36.1 contains bug fixes around WebSocket behavior and browser tool safety checks, plus internal refactors to prepare handwritten runtimes for chunked uploads and vector store polling. No explicit breaking changes are called out in the release notes.

  • v3.36.0

    v3.36.0 adds Vertex AI support for the Responses API and introduces a new Audio API surface for managing voice consents. It also updates admin usage response models to report 30-minute cache write counts, and includes several WebSocket and realtime client robustness changes.

    Features
  • v3.35.1

    v3.35.1 is a small patch release that focuses on a bug fix for beta Realtime websocket byte handling with websockets 13. The only runtime code change is in the beta realtime `recv_bytes()` implementation, with additional test coverage updates around websocket redirect behavior.

  • v3.35.0

    v3.35.0 adds typed Live sideband server event models, including reflected input and output audio plus transport notifications. It also extends Agent service tier typing to include a new `ultrafast` tier, and fixes realtime WebSocket byte handling for websockets 13.

    Features
  • v3.34.0

    v3.34.0 adds two outward-facing API capabilities: a new Realtime Translation WebSocket sessions feature, and a new degraded external storage status value. The code diff also introduces a new Translation session helper implementation plus additional WebSocket internals updates that are not covered by the release notes.

    Features
  • v3.33.0

    v3.33.0 adds support for safety identifiers on image generation and editing requests, and surfaces the effective safety identifier on image-related response models and events. It also introduces a shared typed definition for the Decisions prompt_cache_breakpoint parameter and updates the input parameter types to reference it.

    Features
  • v3.32.0

    v3.32.0 adds new API capabilities including prompt-based voice creation, prompt caching options, and new Live session features for audio fixture playback and Decisions. It also improves realtime/error diagnostics and refactors websocket connection handling to share logic across clients. The diff shows multiple API surface and typing changes beyond what the release notes explicitly spell out.

    Features
  • v3.31.0

    v3.31.0 introduces the first set of new browser and computer toolset support in the SDK (labeled as part 1 of 4). The diff primarily adds new documentation and example/demo code for browser automation and integrates Playwright-based type checking and security checks into CI.

    Features