Cerebras Systems priced its initial public offering at $185 per share Wednesday, raising $5.1 billion in the largest AI-hardware offering since the current cycle began. The AI chipmaker sold 27.6 million shares, exceeding the 24 million initially planned. Shares opened at $221, a 19.5% first-day gain, and closed at $214, valuing the company at $16.8 billion fully diluted.
The company designs wafer-scale processors that compete with Nvidia's datacenter GPUs on large-language-model training workloads. Its CS-3 chip contains 900,000 cores on a single wafer, compared to the modular approach every other AI-compute vendor uses. The IPO follows $715 million in trailing revenue and a net loss of $127 million in the twelve months ending June. Lead underwriters Morgan Stanley and Goldman Sachs priced the deal at the high end of the $170-$185 range after extending the roadshow two additional days to accommodate demand.
The pricing matters because it establishes a public-market clearing price for non-Nvidia AI-compute exposure at meaningful scale. Cerebras competes on throughput for specific training tasks, not inference or general-purpose workloads, which means buyers are underwriting a narrower wedge of the AI-infrastructure stack. The valuation represents 33x forward revenue on $7.2 billion in projected bookings, a premium to AMD's 21x but a discount to Nvidia's 38x. Allocators willing to pay that multiple are signaling belief that the AI-compute market supports at least two architectural approaches at scale, which has second-order implications for hyperscaler capex diversification and merchant-silicon roadmaps through 2026.
The debut also confirms that public markets will pay for growth in this category even when profitability remains distant. Cerebras guided to $1.1 billion in revenue for fiscal 2025, a 54% increase, with EBITDA breakeven expected in fiscal 2026. The company's largest customer, G42 in Abu Dhabi, represented 83% of revenue in the most recent quarter, a concentration risk the S-1 disclosed but the market discounted. That customer concentration typically compresses valuations in enterprise-software IPOs, but the AI-hardware label appears to override standard risk-discount frameworks, at least in this window.
Watch whether Cerebras maintains its $214 close through the first quarterly earnings call, expected in early December. That will reveal whether the initial pop was index-fund rebalancing or genuine accumulation by long-duration tech funds. Also watch for G42 revenue mix in the Q3 print; any decline below 75% would signal customer diversification and likely trigger a re-rating higher. The lock-up expires 180 days from today, putting insider sales in early May 2025, which coincides with the typical window for follow-on offerings in this category.
The deal prices one day after the Federal Reserve's 50-basis-point rate cut, which compressed the risk-free rate for long-duration growth bets by the exact margin that makes this valuation mathematically defensible at 33x forward revenue instead of 28x.
The takeaway
Cerebras priced at **$185**, raised **$5.1 billion**, and closed up **16%**, confirming public markets will pay premiums for non-Nvidia AI-compute exposure.
Open a Brand101 Brand Room — the standard in corporate identity. Or shop the full 70K catalog and virtually proof any product right now. Or talk to Celeste for the fast quote. Or route through the named-account desk.
Two hundred brands. Eight months in hand. $0.003 per impression.
The branded-identity layer Chiefs of Staff and heritage CMOs route through. Already imprinting for Nike, YETI, Patagonia, Thule, Stanley, Moleskine, and one hundred and ninety-five more. Five intelligence desks on the morning reading list of the operators who sign the invoices.
$0.003per impression · vs Meta 0.007 CPM
8 monthsretention in hand · vs Meta 0.8 seconds
200brands you already own · Nike · YETI · Patagonia
Twenty-four AI workers. Seven hundred branded videos live. 24/7.
Celeste and Sora hold conversations. Cleo renders twenty videos per run. Vivienne distributes them across LinkedIn, X, Bluesky, Substack. The MCP catalog routes AI agents straight into the quote flow. The House runs on its own AI stack — two dozen workers operating continuously.
Seventy thousand products. Two hundred brands. One press room.
Own facilities in Virginia Beach. Short-run from twenty-five units, volume to five hundred thousand. Two hundred authorized national brands, seventy thousand SKUs with virtual proofing on every one. Art archived for reorders. Net-thirty corporate terms, NDA-standard white-label.
Full-service agency. AI-native. Five desks in-house.
Huang Goodman: strategy, positioning, identity, creative, messaging, AI-system integration. Media operations across LinkedIn, X, Bluesky, Substack, ChatGPT. For principals building the operating layer their household and portfolio run on.
A single point of contact. Quiet delivery. The file stays on the desk between engagements. Programs for single-family offices, heritage-house CMOs, sports-team ownership groups, and the agencies that route through us for production.
SFO · Chief of Staff desk. Principal household, properties, aircraft, yacht, calendar, philanthropy — one file.
Shop seventy thousand products. Virtual proof on every one. 24/7.
Drop your logo on any product and see the virtual proof before asking. Quote routes direct to the desk. MCP catalog for AI agents. Celeste for the fast conversation. Full self-service checkout in development.