OpenAI Previews Ultrafast Mode for 14-Fold GPT-5.6 Sol Speedup

OpenAI positioned GPT-5.6 Sol, its flagship model for complex professional work, as the highest-capability tier in a family that also includes Terra and Luna. Standard Sol API pricing is $5 per million input tokens and $30 per million output tokens. Ultrafast matters because real-time AI has typically required smaller or specialized models; pairing frontier intelligence with very low latency could expand uses in incident response, financial research, voice support and interactive experimentation.

On Aug. 13, 2026, OpenAI previewed Ultrafast, a new API service tier powered by Cerebras’ wafer-scale computing technology. The company said it can run GPT-5.6 Sol as much as 14 times faster than Standard processing and generate up to 750 output tokens per second. Access is initially limited to a select group of customers, with capacity set to expand over time. OpenAI did not disclose pricing or a date for broader availability.