AI Speed Takes Center Stage as OpenAI and Google Push Faster Models

OpenAI and Google have introduced faster artificial intelligence models that prioritize speed as a key feature. OpenAI's new tier, called Ultrafast, runs its GPT-5.6 Sol model up to 14 times faster than the standard tier, at up to 750 output tokens per second. The company claims this is a shift in AI pricing, where businesses will pay for speed separately from capability.

The Ultrafast tier is currently available as a preview to a small group of customers and includes companies such as Jane Street, Podium, Basis, and Rogo. These early adopters are testing the faster model for various tasks including coding, financial research, customer support, voice applications, and commerce.

Google has also launched its own faster AI model, Gemini 3.7 Flash, which costs $0.75 per million input tokens and $3.75 per million output tokens through December 31. The price will double on January 1, 2027, to the previous level of Gemini 3.6 Flash.

The shift towards prioritizing speed in AI pricing is expected to split business spending into two lanes: fast and expensive AI for jobs that require real-time decisions, and slow and cheaper AI for tasks where speed doesn't matter as much.