OpenAI's GPT-5 Pro and Sora 2 aims to strengthen its developer ecosystem

OpenAI

OpenAI has expanded its developer ecosystem with several major product launches, including GPT-5 Pro, a new video generation model called Sora 2, and a smaller, more affordable voice model designed for real-time use.

The updates were unveiled at OpenAI’s Dev Day and reflect the company’s intent to strengthen its developer community by providing more powerful and cost-efficient APIs. GPT-5 Pro is tailored for high-precision use cases in finance, healthcare, and legal sectors, where accuracy and depth of reasoning are essential.

Sora 2, the upgraded version of OpenAI’s video and audio generator, is now available in API preview. It can create realistic, physically consistent scenes with synchronised sound, ambient effects, and creative flexibility over visual styles and camera movements. Developers can use it to design ads, product visuals, or creative concepts, while companies like Mattel are already experimenting with it for toy design.

OpenAI’s new voice model, gpt-realtime mini, is 70% cheaper than its predecessor while maintaining the same audio expressiveness and quality. It enables low-latency streaming for real-time voice interactions — a crucial step as AI voice interfaces gain traction globally.

Altman said these tools will help developers build smarter, more interactive products and hinted that AI voice will soon become a primary mode of human-computer interaction.

By merging its new models with app integration capabilities inside ChatGPT, OpenAI is positioning itself as the core platform for building intelligent, multi-modal experiences. The company’s developer-first strategy shows how it plans to retain dominance in the increasingly crowded AI space.