OpenAI previews GPT-5.6 Sol Ultrafast at 750 tokens per second

OpenAI is previewing GPT-5.6 Sol Ultrafast, an API service tier that uses Cerebras hardware to deliver up to 750 output tokens per second—14 times faster than Standard processing. The limited rollout targets latency-sensitive work such as incident response, where teams need to analyze logs, traces, code changes, and reports while an outage is still developing. OpenAI has not disclosed separate pricing for Ultrafast; the standard GPT-5.6 Sol API rate is listed at $5 per million input tokens and $30 per million output tokens. Source

Läs hela artikeln hos källan.