{"id":30429,"date":"2025-10-07T10:59:39","date_gmt":"2025-10-07T10:59:39","guid":{"rendered":"https:\/\/metaverseplanet.net\/blog\/?p=30429"},"modified":"2026-01-05T09:01:53","modified_gmt":"2026-01-05T09:01:53","slug":"openai-to-developers-sora-2-and-gpt-5-pro-coming-to-the-api","status":"publish","type":"post","link":"https:\/\/metaverseplanet.net\/blog\/openai-to-developers-sora-2-and-gpt-5-pro-coming-to-the-api\/","title":{"rendered":"OpenAI to Developers: Sora 2 and GPT-5 Pro Coming to the API"},"content":{"rendered":"\n

The leading artificial intelligence company OpenAI<\/strong> announced significant updates for its API<\/strong> ecosystem during its Dev Day<\/strong> event. The company added the GPT-5 Pro<\/strong>, Sora 2<\/a><\/em><\/strong>, and gpt-realtime mini<\/strong> models to its API.<\/p>\n\n\n\n

At its annual Dev Day<\/strong> event for developers, OpenAI introduced major innovations to its API<\/strong> ecosystem. The company announced the new GPT-5 Pro<\/strong> language model, the advanced video generation system Sora 2<\/strong>, and a new low-latency audio model, initiating a major expansion in the field of generative AI<\/strong>.<\/p>\n\n\n\n

The new models are seen as part of OpenAI’s goal to empower its developer community. The company also announced its own agent creation tool<\/strong> and the ability to develop applications directly within ChatGPT<\/a><\/em><\/strong>.<\/p>\n\n\n\n

\n\n\n\n

New Models Added to the OpenAI API<\/h2>\n\n\n\n
\"\"<\/figure>\n\n\n\n

Speaking at the event, CEO Sam Altman<\/strong> emphasized that GPT-5 Pro<\/strong> is specifically designed for developers working in fields requiring high accuracy and deep reasoning, such as finance, law, and healthcare. The model offers an expansive 400,000-token context window<\/strong>, ensuring consistency in long texts. Pricing is set at $15 per 1 million input tokens<\/strong> and $120 per 1 million output tokens<\/strong>.<\/p>\n\n\n\n

OpenAI also noted that voice interactions will be one of the primary ways to communicate with AI in the future. In line with this, the company made the new low-cost, high-speed audio model, “gpt-realtime mini,”<\/strong> available via the API. The new model is 70% cheaper<\/strong> than OpenAI’s previous advanced audio model while maintaining the same audio quality and expressive power. This system supports real-time conversational interactions with low-latency audio streaming<\/strong>.<\/p>\n\n\n\n

In addition to the language and audio models, developers now have access to OpenAI’s latest video generation system, Sora 2<\/strong>. The model, which was announced alongside the recently introduced Sora<\/strong> application, is offered as an API preview<\/strong>. This platform, featuring TikTok-like short videos<\/strong>, allows users to generate videos of themselves, their friends, or imaginary scenes simply by typing a text command. Sora 2 API<\/strong> pricing starts at $0.10 for 1 second<\/strong> and $1 for 10 seconds<\/strong>. Sora 2 Pro<\/strong> starts at $0.30<\/strong>.<\/p>\n\n\n\n

Sora 2<\/strong> offers more realistic, physically consistent scenes, synchronized sound, and greater creative control compared to its previous version. Users can now direct scenes in detail, from camera movements to visual styles. Altman summarized the model’s visual capabilities by saying, “For example, you can take an iPhone shot and tell Sora to turn it into a sweeping, cinematic wide shot.” He also noted that Sora 2<\/strong> can generate not only speech but also ambient sounds<\/strong>, synchronized effects<\/strong>, and audio landscapes<\/strong> based on the visual context.<\/p>\n\n\n\n

You Might Also Like;<\/h3>\n\n\n