OpenAI Announces O3 and O4-Mini Models Before GPT-5 Release

OpenAI, a leading artificial intelligence company, has announced a significant change in its release plans. CEO Sam Altman revealed that the company will release two intermediate models, O3 and O4-Mini, before the highly anticipated GPT-5. This decision comes as OpenAI faces technical challenges in integrating all components into GPT-5, which is expected to be a fully multimodal model. The staggered release of O3 and O4-Mini is aimed at ensuring sufficient capacity to support the unprecedented demand for GPT-5.

Altman's announcement on X highlighted that the release of these intermediate models will allow OpenAI to enhance GPT-5 beyond its original capabilities. The new model is expected to merge all of OpenAI's specialized models into a single system, eliminating the need for ChatGPT to switch between different models based on the prompt. This unified model will handle reasoning, standard language processing, and image generation functions seamlessly.

The technical specifications for O3 and O4-Mini remain undisclosed, but industry experts anticipate that these models will bridge the capabilities gap between GPT-4 and GPT-5. GPT-5 is expected to feature substantial improvements in reasoning, planning, and memory functions, making it a significant advancement in AI technology.

OpenAI's latest release, the reasoning-focused O1 Pro, was priced at $150 per million tokens for input and $600 per million tokens generated, which is double the input cost of GPT-4.5 and ten times the price of regular O1. This pricing strategy reflects the advanced capabilities of O1 Pro, which is designed to handle complex reasoning tasks.

The revised roadmap comes just days after OpenAI closed a historic $40 billion funding round, the largest single fundraising event by any private tech company. This funding will support the development and release of the new models, ensuring that OpenAI remains at the forefront of AI innovation.

OpenAI's decision to release O3 and O4-Mini before GPT-5 places the company in an increasingly competitive field of AI heavyweights. Other companies, such as google, DeepSeek, and Grok, are also rolling out advanced models with extended thinking capabilities. This competitive landscape will drive further innovation and improvement in AI technology, benefiting users and industries alike.