OpenAI and Cerebras Preview "Ultrafast" Mode for GPT-5.6 Sol, Running the Flagship at Up to 750 Tokens per Second
A new OpenAI API service tier powered by Cerebras runs the same GPT-5.6 Sol model up to about 14 times faster than Standard, with no reported quality loss — in limited preview.
2026-08-13 · by Moe Ameen
Published 2026-08-13 · Updated
Short answer
OpenAI and Cerebras previewed Ultrafast on August 13, 2026, a new OpenAI API tier that runs GPT-5.6 Sol at up to 750 output tokens per second, up to 14 times faster than Standard. It is the same model on Cerebras wafer-scale chips with 44 GB of on-chip SRAM. Cerebras reported a 5.6x end-to-end speedup on GDP-Val with no quality loss. Access is limited preview only.
Key takeaways
- Ultrafast was previewed August 13, 2026: up to 750 tokens/sec, up to 14x faster than Standard.
- It runs the same GPT-5.6 Sol on Cerebras' Wafer-Scale Engine with 44 GB of SRAM per chip.
- Cerebras reported 11x faster than Claude Fable 5 and 5x faster than Opus 4.8 Fast.
- It is API-first and open only to select customers for now.
- Faster tokens are still just text. Producing and posting the content is the slow part.
What is GPT-5.6 Sol Ultrafast mode?
On August 13, 2026, OpenAI and Cerebras shared an early look at "Ultrafast," a new OpenAI API service tier that runs the flagship GPT-5.6 Sol model at up to 750 output tokens per second — which OpenAI describes as up to roughly 14 times faster than Sol's Standard processing. The key claim is that speed comes with no intelligence trade: it is the same GPT-5.6 Sol, and on the GDP-Val benchmark the companies reported a 5.6x end-to-end speedup with no measured quality degradation.
The acceleration is a hardware story. Cerebras serves Sol on its Wafer-Scale Engine, whose wafer-sized chips hold model weights on-chip in 44 GB of SRAM, avoiding the memory-bandwidth bottleneck that limits frontier-model inference speed on conventional GPUs.
In the companies' own comparisons, Ultrafast runs about 5x faster than Claude Opus 4.8 on Fast mode and about 11x faster than Claude Fable 5; on the Humanity's Last Exam benchmark it completed 2,500 questions in 11 hours and 11 minutes, against roughly 78 hours for Fable 5. Cerebras frames the target work as latency-bound and high-stakes: legal briefs, financial models, engineering reports, production-outage diagnosis, cybersecurity response, and long agent chains.
The important caveats are access and scope. Ultrafast launches first in the OpenAI API and is in limited preview — available initially to a select group of customers, with access expanding over time — so most people cannot use it yet. And it is a speed tier, not a capability change: GPT-5.6 Sol still outputs text and code and reads images. It renders no images, video, or audio and publishes nothing. As always with a preview, treat the specific throughput and benchmark figures as a snapshot of the announcement.
Why does an ultrafast model tier matter for creators?
- The bottleneck in an AI workflow is shifting from "waiting on the model" to everything after it — near-instant reasoning does not shorten producing and publishing.
- Frontier intelligence at interactive speed makes high-volume drafting practical: a month of hooks, scripts, or angles in one sitting instead of across days.
- It is preview-gated and API-first, so it is a developer capability today, not a click-to-use feature — a gap between "in the news" and "in your hands" worth explaining to an audience.
- Speed is now a competitive axis between frontier models, not just quality — but for content, faster tokens still are not a finished, scheduled post.
- Specialized inference hardware (Cerebras' wafer-scale chips) underpinning a frontier model is itself a timely, high-search topic your audience is asking about this week.
How do you act on Ultrafast news with Kompozy?
The fastest thing you can do with this news is not use Ultrafast — it is publish a clear take on it before everyone else. "OpenAI just made GPT-5.6 Sol 14x faster on Cerebras" is exactly the timely, high-intent topic your audience is searching this week, and most creators will still be waiting for preview access while the interest peaks. Drop your angle into Kompozy and it fans one point of view into a blog explainer, a carousel that breaks down what a speed tier actually is, short captioned clips, a quote graphic on the headline number, and platform-native posts — all in your voice through the Persona Brief — then schedules and publishes them across Instagram, TikTok, YouTube, LinkedIn, X, Facebook, Pinterest, and Threads plus a blog and newsletter. Being early and useful on a story like this is how one take becomes a week of content.
There is also a workflow read worth keeping. If a faster model tempts you to wire your pipeline to whichever endpoint is quickest this month, resist it — the fastest tier changes constantly, and token speed is not speed to a published post. Kompozy runs generation on managed Claude and OpenAI models with the model layer abstracted away, so you never re-wire when a new fast tier lands; you set a Persona Brief and approve outputs while the engine renders persona video, carousels, images, blogs, and newsletters and ships them on a schedule. Let Ultrafast (or any fast model) draft upstream if you have access; let Kompozy own the producing-and-publishing half that no inference speedup touches.
What are the quick takeaways?
- OpenAI and Cerebras previewed Ultrafast mode for GPT-5.6 Sol on August 13, 2026 — up to 750 output tokens/sec, ~14x faster than Standard, same model.
- The speed is from Cerebras' Wafer-Scale Engine (44 GB on-chip SRAM); the companies reported no quality loss and a 5.6x GDP-Val speedup.
- It is API-first and in limited preview to select customers, so broad access is not available yet.
- A faster model produces more text, not finished content — Kompozy turns drafts into 20 formats published across nine platforms, whatever model runs upstream.
What does this look like in Kompozy?
A real infographic photo post from the Kompozy workspace. It is a format for turning a headline number like this one into a post.
Who is this for, and who should skip it?
Best for:
- Developers with latency-bound work like outage response or long agent chains
- Enterprises already in the OpenAI API preview program
- Creators covering AI infrastructure news this week
Not ideal for:
- Most people today, since access is limited to select customers
- Anyone expecting smarter answers, since it is the same model, only faster
- Creators who want images, video or publishing, which Sol does not do
Frequently asked questions
What is GPT-5.6 Sol Ultrafast mode?
It is a new OpenAI API service tier, previewed with Cerebras on August 13, 2026, that runs the existing GPT-5.6 Sol model at up to 750 output tokens per second — roughly 14x faster than Sol Standard. The intelligence is unchanged; only the speed is different, powered by Cerebras' wafer-scale hardware.
Can I use Ultrafast mode right now?
Not broadly. At the August 13, 2026 preview it is in limited availability to a select group of customers, with access expanding over time, and it launches first in the OpenAI API — so it is developer-facing rather than a consumer feature you can click on today.
Does Ultrafast change what GPT-5.6 Sol can produce?
No. Ultrafast only makes Sol answer faster. It still outputs text and code and reads images — it renders no video, images, or audio and publishes nothing. Faster drafting sits upstream of content; turning drafts into finished, scheduled posts is the job of a content engine like Kompozy.
How is this different from GPT-5.6 Sol Ultra?
"Ultrafast" is a speed tier — the same Sol, returned faster, on Cerebras hardware. "Ultra" is a separate mode that uses subagents and a max-reasoning-effort setting for depth on hard tasks, and is being wired into Codex for agentic coding. One is about throughput, the other about how hard the model works on a single problem.
How do we know this?
We checked every figure, date, quote and product claim on this page against the sources listed below on 2026-10-23, and removed or rewrote anything we could not confirm. Speed, hardware and benchmark figures come from Cerebras' post, the 14x comparison and preview status from IT Brief, and the Sol Ultra description from Techsy. Kompozy plan and format details come from kompozy.io.
Sources
- Cerebras — Accelerating GPT-5.6 Sol Ultrafast with OpenAI (read 2026-10-23)
- IT Brief NZ — OpenAI launches Ultrafast GPT-5.6 Sol for businesses (read 2026-10-23)
- Techsy — GPT-5.6 Sol Ultra in Codex (read 2026-10-23)
- Kompozy — Pricing and plans (read 2026-10-23)