OpenAI Unveils Ultrafast GPT-5.6 Sol Powered by Cerebras

OpenAI has officially showcased the fruits of its partnership with Cerebras, introducing a new Ultrafast API tier. This upgrade allows its flagship model, GPT-5.6 Sol, to run up to 14 times faster than normal, reaching speeds of up to 750 tokens per second. The collaboration, which was first revealed in January, leverages Cerebras specialized hardware designed specifically for rapid computation. In practical tests, the Ultrafast tier completed a rigorous 2,500-question exam in just 11 hours, a task that took the competing Fable model 78 hours to finish. Early users and OpenAI staff have reported groundbreaking efficiency gains, noting that tasks like security investigations have been cut from hours to mere minutes. Currently, the Ultrafast tier is available as an invite-only preview, with OpenAI planning to expand access as more computing power becomes available.