OpenAI Releases GPT 4.5; Its Most Expensive Model

OpenAI has officially launched GPT-4.5, its most advanced AI language model. The model is currently in research preview and available to ChatGPT Pro users, offering enhanced writing capabilities, improved knowledge accuracy, and a more natural, less hallucinatory experience.

The model is rolling out in stages, starting with ChatGPT Pro users, followed by Plus and Team users next week. Enterprise and Education users will gain access shortly after.

Sam Altman, CEO of OpenAI, shared the announcement on X, stating, “GPT-4.5 is ready. It’s a giant, expensive model. We wanted to launch it to Plus and Pro at the same time, but we've been growing a lot and are out of GPUs.”

To meet demand, OpenAI plans to add tens of thousands of GPUs next week for the Plus tier, with hundreds of thousands more coming soon, which are expected to be fully utilized.

GPT-4.5, however, is not classified as a reasoning model. It builds on the GPT series with improvements such as pretraining, supervised fine-tuning, and RLHF, setting it apart from models like OpenAI’s O-series.

Pricing for GPT-4.5 includes:

• $75.00 per million input tokens
• $37.50 per million cached input tokens
• $150.00 per million output tokens

Altman added, “This isn’t a reasoning model and won’t crush benchmarks. It’s a different kind of intelligence, and there’s a magic to it I haven’t felt before.”

According to reports from last week, Microsoft engineers are expanding infrastructure capacity in anticipation of the upcoming releases of OpenAI’s latest large language models. After GPT4.5, OpenAI could release GPT-5 in May.