Cerebras (CBRS)
The desk's stance
On our research list, not on the slate.
What the company does
Builds wafer-scale AI chips — instead of cutting a silicon wafer into hundreds of small GPUs, it ships the whole dinner-plate-sized wafer as one giant processor, wired for ultra-fast AI inference.
Why it matters: The radical architecture bet in the designer column: if wafer-scale inference wins even a niche, it chips at Nvidia’s margins from below.
Outlook: Public since its 2025 IPO with wafer-scale inference deals ramping (G42 the anchor customer); the stock trades on the architecture bet more than current earnings.
In short: Wafer-scale AI chips and systems. Category: AI Infrastructure & Wafer Silicon. Sector: Technology, Semiconductors.
G-Street Value and Score
G-Street Value: $278.67, computed Oct 18, 2026 (confidence: low).
How we got there: 100% analyst leg $278.67 (5 analyst targets, credibility- and freshness-weighted, after 8% optimism discount) + factor model can't price this profile (below its floor) — analyst leg only.
G-Street Score: 49 out of 100 (Weak).
- Top-line growth & guidance: 27 of 32
- Margin power: 5 of 15
- Revenue visibility & quality: 6 of 11
- Profitability & balance sheet: 7 of 24
- Capital allocation & governance: 1 of 6
- Valuation, sentiment & momentum: 3 of 12
Our current thesis
Updated Oct 18, 2026
The thesis
- Proprietary wafer-scale architecture delivers industry-leading low-latency inference speeds versus traditional GPU clusters.
- Contracted $25.4B RPO backlog provides multi-year revenue visibility across large-scale enterprise deployments.
- Substantial $7.92B cash reserve buffers aggressive datacenter buildouts and next-generation CS-4 chip manufacturing.
Risks
- Peer NVDA generates 64% net margins, highlighting Cerebras's steep profitability deficit and $450.5M loss.
- Reports of OpenAI deploying Nvidia workloads threaten execution against multi-year compute commitments.
- Forward P/E of 99.9x leaves valuation vulnerable to further datacenter leaseback margin deterioration.
What we're watching
- Q3 2026 earnings report and gross margin trajectory scheduled for November 10, 2026.
- Deployment milestones for the AMD Helios disaggregated solution targeted for Q4 2026.
- Price stabilization above the key 52-week low support level of $126.05.
Latest
CBRS slid 10.5% to $127.21, testing 52-week lows ($126.05) following $387.7M in insider selling and reports of OpenAI tapping Nvidia for workloads. Q2 GAAP gross margin compressed to 14.2% alongside a $450.5M net loss driven by heavy infrastructure leaseback costs. The balance sheet holds $7.92B in cash to fund datacenter capacity ahead of Q3 earnings on November 10, 2026.
Key numbers
As of Oct 16, 2026: market cap $30.2B · revenue growth +74.3% year over year · gross margin 34.9% · net margin -75.3% · forward P/E 99.9 · next earnings Nov 10, 2026.
Analyst price targets
Published analyst targets we could trace to a named source.
- Rosenblatt, Buy: $300 target (August 2026). Source: marketbeat.com
- Wedbush, Outperform: $290 target (August 2026). Source: marketbeat.com
- Morgan Stanley, Overweight: $279 target (August 2026). Source: www.marketbeat.com
- UBS, Buy: $330 target (August 2026). Source: marketbeat.com
- Citigroup, Buy: $320 target (August 2026). Source: marketbeat.com
Decision history
- Jul 15, 2026 — Unpinned at $203.81. Unpinned — back under AI management
On the Gainz Lab maps
Who it does business with
- TSMC supplies Cerebras — Wafer-scale chips (about $1.0B a year). TSMC fabricates Cerebras’ dinner-plate-sized wafer-scale engines.
- Cerebras competes with Nvidia — Wafer-scale vs GPUs. Cerebras attacks AI inference with one giant chip instead of clusters of GPUs — a bet that Nvidia’s networking advantage can be designed away.
In the Daily Brief
Named in 5 editions on 5 days.
- Oct 1, 2026: Yields surged right back to 2002 levels, but OpenAI keeping GPT-6 on our Nvidia makes the macro noise bearable
- Aug 18, 2026: Greed stays high as AI hardware leads our way, and memory chips rally, but Reddit's S&P pop faded quick
- Aug 17, 2026: Oil sent yields soaring, which dragged the big names down, but AI hardware shrugged it off while Reddit shed some weight
- Jul 16, 2026: Yields dipped with CPI, yet our treasured tech found itself swimming against the capital flow
- Jul 15, 2026: Inflation whispers sweet nothings, yet our high-flying AI chips are bracing for export headwinds