---
title: "OpenAI now decides what Cerebras's wafer is worth"
description: "Cerebras proved this month that its wafer serves trillion-parameter models fast, and the model it showcased was open-weights, not OpenAI's. With about $20 billion of $25 billion in contracted business tied to OpenAI and the last lock-ups expiring before the November 12 report, one buyer now sets the wafer's price."
publisher: "The Inference"
section: "Chips"
published: 2026-10-25T13:13:17.777Z
modified: 2026-10-25T13:13:17.777Z
canonical: https://theinference.org/article/openai-now-decides-what-cerebras-s-wafer-is-worth
language: en
keywords: "Cerebras, OpenAI, AI Chips, Inference, AMD"
---

# OpenAI now decides what Cerebras's wafer is worth

> Cerebras proved this month that its wafer serves trillion-parameter models fast, and the model it showcased was open-weights, not OpenAI's. With about $20 billion of $25 billion in contracted business tied to OpenAI and the last lock-ups expiring before the November 12 report, one buyer now sets the wafer's price.

On October 22 Cerebras [published the engineering behind its sales pitch](https://www.cerebras.ai/blog/how-we-deploy-trillion-parameter-frontier-models-on-cerebras): the open-weights Kimi K3 runs across 59 of its dinner-plate wafers, whole layers of the model resident in on-chip memory, only intermediate results moving between wafers [1]. The post reached [Hacker News](https://news.ycombinator.com/item?id=50227356) over the weekend [2]. The telling detail is the model it chose to showcase: open-weights, not [OpenAI](https://theinference.org/markets/companies/openai)'s. The wafer's speed is settled; its price is not, and one customer sets it.

That customer is OpenAI, behind about $20 billion of the $25 billion [Cerebras](https://theinference.org/markets/companies/cerebras-systems) has contracted for future delivery, against [roughly $210 million of core revenue last quarter](https://www.cnbc.com/2026/10/23/investor-paul-meeks-buying-cerebras-cbrs.html) [3][4]. The pairing with [AMD](https://theinference.org/markets/companies/amd), announced in July when AMD chief executive [Lisa Su](https://theinference.org/markets/people/lisa-su) unveiled her Helios server system [5], deepens that dependence rather than easing it.

## A second rival's silicon now sits in front of the wafer

In the joint system, AMD's Helios rack, holding up to 72 MI400-series accelerators, handles prefill, the first pass over a prompt; the Cerebras wafer handles decode, writing the answer token by token, with up to five times better energy efficiency claimed [6]. [Amazon Web Services](https://theinference.org/markets/companies/amazon) already runs the same split with Cerebras, its Trainium chip on prefill and the wafer on decode [3].

That gives Cerebras a second way to sell decode, and OpenAI a second way to buy it. CBRS rose about 4% on Thursday to a one-month high, its fourth straight up session, and still trades roughly 72% below the $386 reached after its May listing [6][3].

## Speed is settled, so the fight has moved to price

The two halves of inference have different physics. Prefill digests a long prompt in parallel, which racks of GPUs already do well; decode emits one token after another and is bound by memory bandwidth, which the wafer's on-chip memory supplies [6][3]. Outside measurement backs the company's method: Artificial Analysis clocked the wafer at 981 output tokens a second on Kimi K2.6, an earlier Kimi release, 6.7 times the next-fastest GPU cloud [7]. Coding agents live on fast decode. On speed, the argument is over.

## One buyer can now buy decode two ways

What is unsettled is price. AMD says OpenAI plans to deploy Helios at scale [6]. [Reports keep asking](https://www.ad-hoc-news.de/boerse/news/unternehmensnachrichten/cerebras-draws-two-kinds-of-pressure-expiring-lock-up-and-openai-hardware-doubts/70355408) whether the "Ultrafast" tier of OpenAI's GPT-6.1 Sol [ran on Nvidia GPUs rather than Cerebras hardware](https://theinference.org/article/altman-calls-cerebras-a-close-partner-as-openai-s-fastest-tier-runs-on-nvidia), and [Microsoft](https://theinference.org/markets/companies/microsoft)'s Maia 300 inference chip is arriving [8]. If OpenAI can put its prompts on an AMD rack and still buy decode behind it, it no longer needs Cerebras's whole stack. A buyer holding two ways to buy pays less than one holding a single supplier.

*Chart: **Cerebras's contracted promises tower over what it bills each quarter** Bar chart comparing Cerebras's total remaining performance obligations of $25.4 billion, the $20 billion of those obligations tied to OpenAI, and last quarter's core revenue of about $0.21 billion.*

*Remaining performance obligations and last quarter's core revenue, US dollars, from company disclosures reported by CNBC and MarketBeat. [3][4]*

## The calendar prices the stock before the earnings do

Nineteen million shares become tradable on October 28 and 14 million more on November 10, after which every share can trade [9]. Insiders, among them the chief technology officer and chief financial officer selling under pre-arranged plans, [took out nearly $388 million across three months](https://theinference.org/article/insiders-sold-387-7-million-of-cerebras-stock-19-4-million-more-shares-unlock-wednesday) [4]. A lock-up expiry makes shares sellable; it does not show they were sold, and several sellers are funds illiquid for a decade taking profit rather than verdicts on the hardware [8][9].

The test is [the third-quarter report on November 12](https://theinference.org/article/cerebras-rallied-9-on-altman-s-praise-but-openai-s-20-billion-revenue-gap-sets-a-november-test) [7]. The number to watch is revenue tied to OpenAI against last quarter's $210 million base [3]. If it is climbing, the backlog is turning real and AMD is a second sales channel. If it is flat, the wafer has a price ceiling that no speed benchmark removes.

## Takeaway

Cerebras settled the speed argument by serving a trillion-parameter open-weights model, but the model it chose was not OpenAI's. With $20 billion of its $25 billion backlog tied to one buyer that can now buy decode two ways, and 33 million shares unlocking before the November 12 report, OpenAI effectively prices the wafer.

## Sources

1. [Cerebras blog, Sarah Chieng, "How we deploy Trillion-parameter frontier models on Cerebras", 22 October 2026](https://www.cerebras.ai/blog/how-we-deploy-trillion-parameter-frontier-models-on-cerebras)
2. [Hacker News, "Cerebras Deploys Trillion-Parameter AI Models", October 2026](https://news.ycombinator.com/item?id=50227356)
3. [CNBC, Paul Meeks, "I warned against buying this AI chip stock. Now I think it could nearly double", 23 October 2026](https://www.cnbc.com/2026/10/23/investor-paul-meeks-buying-cerebras-cbrs.html)
4. [MarketBeat, "Cerebras Systems Inc. $CBRS Shares Newly Acquired by Inspire Investing LLC", 25 October 2026](https://www.marketbeat.com/instant-alerts/filing-cerebras-systems-inc-cbrs-shares-newly-acquired-by-inspire-investing-llc-2026-10-25/)
5. [Business Insider, Geoff Weiss, "AMD takes a shot at Nvidia by betting on AI's next big shift", 23 July 2026](https://www.businessinsider.com/amd-cerebras-partner-ai-inference-helios-system-2026-7)
6. [Stocktwits, "Cerebras Partners With AMD To Deliver Accelerated AI Performance", October 2026](https://stocktwits.com/news-articles/markets/equity/cerebras-partners-with-amd-advancing-ai-2026/cZZnycsR7Uj)
7. [Startup Fortune, Elroy Fernandes, "Cerebras bets on wafer-scale chips as Nvidia's GPUs get scarce and pricey", October 2026](https://startupfortune.com/cerebras-bets-on-wafer-scale-chips-as-nvidias-gpus-get-scarce-and-pricey/)
8. [ad-hoc-news.de, "Cerebras Draws Two Kinds of Pressure: Expiring Lock-Up and OpenAI Hardware Doubts", 23 October 2026](https://www.ad-hoc-news.de/boerse/news/unternehmensnachrichten/cerebras-draws-two-kinds-of-pressure-expiring-lock-up-and-openai-hardware-doubts/70355408)
9. [Brownstone Research, Jeff Brown, "The Largest Merger in History" (Cerebras lock-up Q&A), 23 October 2026](https://www.brownstoneresearch.com/bleeding-edge/the-largest-merger-in-history/)


---

OpenAI now decides what Cerebras's wafer is worth — The Inference. Canonical: https://theinference.org/article/openai-now-decides-what-cerebras-s-wafer-is-worth