Aug 13, 2026
Codex previews an “Ultrafast” tier — GPT-5.6 Sol at up to 14x speed
OpenAI began a limited preview of an Ultrafast service tier running GPT-5.6 Sol at up to ~14x speed (~750 tokens/sec) on Cerebras hardware. No pricing or GA date announced yet.
What changed
OpenAI opened a limited preview of an “Ultrafast” service tier that runs GPT-5.6 Sol on Cerebras inference hardware at up to roughly 14x the usual speed — around 750 output tokens per second. It targets latency-sensitive agent loops where waiting on the model is the bottleneck. As a preview it has no published price, no general-availability date, and it is an API service tier rather than a switch inside the Codex UI yet.
Key details
Model
GPT-5.6 Sol
Speed
up to ~14x (~750 tok/s)
Hardware
Cerebras
Status
Limited preview
Price
Not announced
Availability
API service tier
Official source
openai.com
Read the source →
Full profile: OpenAI Codex →