Cerebras Powers OpenAI API Ultrafast Tier, Running GPT-5.6 at 750 Tokens/sec
Sethwinterroth · x · 2026-08-14
Cerebras announced that it now powers Ultrafast, a new service tier in OpenAI's API, running GPT-5.6 Sol at up to 750 output tokens per second.
This integration creates an entirely new point on the speed-intelligence frontier, enabling enterprises to build with frontier intelligence at real-time speeds. The service is designed to support faster coding agents, research systems that analyze information in seconds, and enterprise agents that reason and act without latency. GPT-5.6 Sol Ultrafast is currently beginning with a limited API preview, with capacity expected to grow over time.
Related event: OpenAI and Cerebras Preview GPT-5.6 Sol Ultrafast Mode at 750 Tokens/s(9 posts)→
More from Infra
- 4x Faster Kimi-K3 Decoding: Red Hat Releases DSpark Speculator — woosuk_k · 2026-08-14
- W&B Crosses 1 Billion AI Training Runs After 9 Years — morgymcg · 2026-08-14
- Nvidia Partners with Finance Giants to Mobilize $500B, Turning AI Compute into an Asset Class — sanjaykalra · 2026-08-14
- NVIDIA's NeMo Switchyard: Model Routing as the Agent Budget Manager — krishnan · 2026-08-14
- Nvidia and Wall Street Aim to Mobilize $500B: Is Compute Really an Asset Class? — whurley · 2026-08-14
- Running MiniMax H3 on RTX 5060 Ti: Resolution is the Real Bottleneck — danielcar · 2026-08-14