OpenAI Python SDK v2.23.0 adds GPT-4 Realtime 1.5 and Audio 1.5 model support
OpenAI released Python SDK version 2.23.0 on February 24, 2026, adding support for two new model options in realtime API calls: gpt-realtime-1.5 and gpt-audio-1.5. The minor update primarily focuses on expanding model availability for real-time applications.
OpenAI Python SDK v2.23.0 Released
OpenAI released version 2.23.0 of its Python SDK on February 24, 2026, introducing support for two new model variants in realtime API operations.
New Models Available
The release adds gpt-realtime-1.5 and gpt-audio-1.5 as available model options for realtime calls. These additions expand developer access to OpenAI's latest real-time capable models through the Python SDK.
The specific capabilities, pricing, and context window details for these models were not disclosed in the SDK release notes. Developers will need to consult OpenAI's API documentation for full technical specifications.
Changes Summary
Features:
- Added gpt-realtime-1.5 and gpt-audio-1.5 model options to realtime calls
Maintenance:
- Improved test resilience for proxy environment variables
What This Means
This update primarily serves as infrastructure expansion rather than a breaking change. Developers using the Python SDK can now specify these newer model variants when building real-time applications. The inclusion of both realtime and audio-specific model options suggests OpenAI is differentiating between audio streaming capabilities and general real-time interaction patterns.
Minor version increments like this are typical for SDK releases that add new model availability without changing API behavior or introducing breaking changes.
Related Articles
Vercel's AI SDK Adds Support for Unannounced OpenAI Model 'GPT-6.1 Sol'
Vercel released @ai-sdk/openai version 3.0.121, a patch that adds SDK support for a model string called 'GPT-6.1 Sol.' OpenAI has not publicly announced or confirmed this model, and no specs, pricing, or benchmarks are available.
OpenAI publishes 372 AI-generated math results on GitHub, claims they solve or advance open problems
OpenAI has published 372 mathematical results generated by an unnamed internal frontier model, hosted on GitHub instead of in peer-reviewed journals. The company claims each result solves or substantially advances an open problem, at an average of about three hours of ChatGPT Pro Thinking compute per result. Many include Lean formalizations, but independent validation of significance is still pending.
OpenAI ships opt-in textGrain text watermarking in API, with EU ChatGPT and Codex rollout to follow
OpenAI has launched textGrain, an invisible statistical text watermark, as opt-in for API customers worldwide on select models. ChatGPT and Codex output in the EU will be watermarked in the coming weeks, in response to the EU AI Act. OpenAI says detection drops from about 92% to 17% when 25% of words in a 400-token passage are replaced.
OpenAI to watermark ChatGPT and Codex text in the EU under AI Act; API opt-in available worldwide
OpenAI will add an invisible watermark to text generated by ChatGPT and Codex in the European Union to comply with the EU AI Act's transparency rules. Developers anywhere can enable it on select API models starting today, but it is off by default. OpenAI's own tests show detection falling from about 92% to 66% after 10% of words are replaced with synonyms.
Comments
Loading...