GPT-5.6 Sol: Ultrafast from Cerebras — 750 Tokens per Second
OpenAI has launched Ultrafast mode for GPT-5.6 Sol — 750 tokens/sec, 14× faster than standard. Behind it is Cerebras, backed by a $10B partnership deal.
OpenAI has launched Ultrafast mode for GPT-5.6 Sol, the flagship model in the series. Output speed reaches 750 tokens per second — 14 times faster than standard mode. The infrastructure is powered by Cerebras Systems, which signed a $10 billion partnership agreement with OpenAI earlier in 2026.
OpenAI is building a three-tier speed monetization model: standard, Fast (up to 2.5×), and Ultrafast (up to 14×, highest price). Ultrafast is currently in preview — available via API and a waitlist for select customers.
The declared use cases are real-time tasks: stock analysis, responding to production incidents, live customer support, and interactive research that previously had to wait in overnight batch queues.
Source: the-decoder.com
Free course
Stop reading about AI — start building with it
The free Claude Code course: your first site, tool or game — no coding. No upsells, no cross-sells — nothing to buy here.
Start free →
Author
Evgenii Arsentev
PhD · Chief Executive Officer, digital health
Articles · Latest articles