github OpenAI Agents SDK · 1h ago · agents

OpenAI Python SDK v0.21.0 adds provider-neutral testing APIs and OpenAI v3 compatibility

from English

OpenAI released version 0.21.0 of the openai-agents-python SDK, introducing new provider-neutral testing utilities and updating compatibility for the OpenAI Python library version 3.

Added agents.testing, agents.realtime.testing, and agents.voice.testing modules to enable deterministic workflow tests without making provider requests.
Updated OpenAI provider compatibility to require openai>=3.0.0,<4, including support for HTTPX2-aware request handling.
Hardened RunState interruption snapshots, recursive agent-tool approvals, max-turn finalization, streaming cleanup, and sensitive-error redaction.
Improved MCP lifecycle snapshot isolation and added configurable retry backoff ceilings.
Added Sandbox Runloop existing-secret support and tightened view-image path grants.
Enforced stricter Voice validation for invalid channels, frame rates, incomplete multichannel frames, and non-finite audio rates.

This release allows developers to test Agent, Sandbox, Realtime, and Voice workflows deterministically while ensuring the SDK remains compatible with the latest OpenAI Python client.

Importance 1/3 Trust 2/3 OpenAI Agents SDK OpenAI AI agents API & product launches

Related articles

media TLDR AI · 11h ago · 11 views

OpenAI previews GPT-5.6 Sol Ultrafast mode; Google releases Gemini 3.7 Flash

OpenAI has previewed Ultrafast, a GPT-5.6 Sol mode capable of generating up to 750 output tokens per second and operating at up to 14x standard processing speed. The approach aims to deliver real-time performance without switching to a smaller model.

media AI News (smol.ai) · 1d ago · 9 views

Google releases Gemini 3.7 Flash; DeepSeek open-sources Harness

Google shipped Gemini 3.7 Flash, its new mid-tier model for coding and agentic workflows, featuring a 50% introductory price cut and significant benchmark gains over version 3.6. The release positions the model as a workhorse for development, with immediate integration across the Gemini API, Android Studio, and external tooling like Cline.

lab OpenAI News · 1d ago · 11 views

OpenAI releases GPT-5.6 for affordable, high-performance agentic workflows

OpenAI has launched the GPT-5.6 model family, designed to deliver frontier-level agent performance at significantly lower costs by optimizing reasoning effort and introducing new API controls. The update allows startups to build faster, more capable agents using smaller models like Luna and Terra for high-volume tasks while reserving larger models for complex judgment.

lab OpenAI News · 3d ago · 15 views

OpenAI reports widening gap as enterprise AI shifts from assistance to execution

Two new OpenAI studies reveal that frontier firms—those in the top 10% of monthly AI usage—are generating 8.3 times more output tokens per active user than typical firms, up from 2.6 times in January. This widening gap reflects a shift in enterprise adoption from using AI for assistance to delegating substantive work to agentic tools like Codex and ChatGPT Work.

lab OpenAI News · 8d ago · 10 views

OpenAI reveals global ChatGPT adoption and usage trends

New data from OpenAI illustrates how users around the world are utilizing ChatGPT, providing country-level insights into adoption rates and evolving behavioral patterns.