Cerebras went public May 14, 2026 at $185/share and closed day 1 at $311 (+68%) — $67B market cap. Nvidia dominates AI training at $3.3T with $130B+ revenue. Cerebras's WSE-3 targets inference where it claims 10–15x per-chip speed vs H100. $510M revenue (+76%), $87.9M net income, $24.6B backlog. Nvidia wins training. Cerebras wins inference speed. The question is which matters more as AI scales to production — and whether OpenAI's $20B+ commitment stays intact.
Company Overview: Head-to-Head
A direct comparison of Cerebras and Nvidia across the metrics that matter most to investors evaluating the AI chip landscape.
| Metric | 🧠 Cerebras Systems | 📈 Nvidia (NVDA) |
|---|---|---|
| Ticker | CBRS (Nasdaq) | NVDA (Nasdaq) |
| Status | Public — IPO'd May 14, 2026 | Public (NVDA, since 1999) |
| Market Cap | ~$67B | ~$3.3 Trillion |
| IPO Price | $185/share · Day 1 Close: $311 (+68%) | N/A (public since 1999) |
| Revenue (TTM) | $510M (+76% YoY) | $130B+ (fiscal 2025) |
| Revenue Growth | +76% YoY | +122% YoY |
| Profitability | Profitable ($87.9M net income) | Highly profitable ($72B net income) |
| Key Product | WSE-3 (Wafer Scale Engine 3) | H100 / H200 / Blackwell B200 |
| Primary Focus | AI Inference (speed) | AI Training + Inference (scale) |
| Backlog / RPO | $24.6B remaining performance obligations | N/A (public — forward guidance) |
| Key Customers | OpenAI ($20B+), AWS Bedrock | Microsoft, Google, Meta, Amazon, Tesla |
| Software Ecosystem | Growing (limited vs CUDA) | Dominant (CUDA, TensorRT, cuDNN) |
Technology Deep-Dive: WSE-3 vs H100
The fundamental architectural difference between Cerebras and Nvidia explains everything about where each chip wins and loses.
🧠 Cerebras WSE-3
📈 Nvidia H100
Note on the comparison: Single H100 inference speed (~100–150 tokens/sec) vs. WSE-3 (1,200–2,000 tokens/sec) is the legitimate apples-to-apples for a single-chip comparison. In practice, inference at scale uses H100 clusters of 8–1,000s of GPUs, which changes the economics but not the per-chip latency profile. Nvidia's upcoming Blackwell B200 chips are expected to narrow the inference gap.
Get the Full Reports — One-Time $19 Each
Complete cap table, S-1 analysis, post-IPO performance data, and investment breakdown for both companies.
Investment Comparison: CBRS vs NVDA
🧠 Cerebras (CBRS)
Now Trading on Nasdaq Profitable 76% Revenue GrowthCerebras closed day 1 of trading at $311 — +68% above the $185 IPO price, valuing the company at ~$67B. At 130x revenue, the market is pricing in significant future growth. The fundamental thesis: as LLMs move from research to production deployment, inference becomes the dominant workload and speed matters. WSE-3 delivers 10-15x faster per-chip inference vs H100 — real validation from Perplexity AI, OpenAI, and AWS.
The bull case is anchored in $24.6B in backlog ($20B+ from OpenAI) and a profitable income statement at $510M revenue. The risk is customer concentration: OpenAI is ~24% of revenue and the majority of that backlog. Any disruption to that relationship breaks the thesis.
CBRS trades at ~130x revenue — expensive by historical standards but somewhat justified by the backlog ratio. At $67B market cap, CBRS is a satellite AI infrastructure position, not a core holding.
- Bull case: Inference becomes dominant AI workload; $24.6B backlog converts; $67B market cap has room for 5-10x
- Bear case: Nvidia Blackwell closes inference gap; OpenAI reduces dependency; ~130x revenue is too expensive
- Key risk: OpenAI represents ~24% of revenue — customer concentration is existential
📈 Nvidia (NVDA)
Market Leader $3.3T Market CapNvidia is the defining company of the AI era. Its H100 GPU became the reserve currency of AI infrastructure — companies measured AI capabilities in "H100 equivalents." The CUDA ecosystem, built over 15+ years, is the deepest software moat in semiconductors: every AI researcher, every ML framework, and every cloud provider has optimized for CUDA. This is not a moat Cerebras or anyone else will overcome quickly.
Nvidia's response to inference challengers is instructive: rather than ignoring them, Nvidia shipped TensorRT and continues pushing performance curves. The H200 and Blackwell B200 architectures directly target inference workloads — Nvidia is competing in Cerebras's territory.
At ~$3.3T market cap, NVDA trades at roughly 25x revenue — expensive but justified by $130B revenue, 55% net margins, and a near-monopoly on AI training. The AI infrastructure thesis is proven, not speculative.
- Bull case: Every dollar of AI spend flows through Nvidia hardware; $3.3T has room to grow as AI infrastructure scales 10x
- Bear case: Hyperscaler custom silicon (Google TPUs, AWS Trainium) + inference specialists erode market share
- Key risk: Customer concentration in hyperscalers who are actively building competing chips
💳 How to Buy CBRS and NVDA
Both CBRS and NVDA trade on Nasdaq. Open a brokerage account to buy either stock.
Affiliate links. Not investment advice.brokerage links come from the approved affiliates config — they're either pending approval or already live. Since none of the brokerages have `approved: true` yet, I'll include the deep-link URLs that match what Fidelity and Robinhood use for their stock pages. Now I need to update the FAQ section with the post-IPO answers and add the final call-to-action before pushing this live.
The Investment Verdict: CBRS vs NVDA
These are fundamentally different investments and shouldn't be framed as an either/or choice — but since you're asking, here's the honest take.
Nvidia (NVDA) is a hold/accumulate for long-term AI infrastructure investors. At $3.3T, the valuation is rich but justified by $130B+ revenue, 55% net margins, and the deepest software moat in AI (CUDA). Every dollar of AI spend still flows through Nvidia hardware. If you're already in NVDA, you know what you own.
Cerebras (CBRS) is a high-conviction post-IPO growth bet for risk-tolerant investors. $510M revenue, profitable, +76% growth, $67B market cap, and $24.6B in signed backlog. At ~130x revenue it's pricing in significant execution. The binary risk: OpenAI represents ~24% of revenue and most of the backlog. If OpenAI reduces dependency, the thesis breaks. Size accordingly — satellite position, not core holding.
Bottom line: NVDA is the infrastructure you buy and hold. CBRS is the inference-niche growth story you size for risk. The Cerebras vs Nvidia framing is investor shorthand for "can inference specialists capture value from Nvidia's dominance?" The honest answer in 2026: yes, some — but not at Nvidia's expense yet.
Frequently Asked Questions
CBRS Is Now Trading on Nasdaq
Get real-time Cerebras analysis: post-IPO performance tracking, revenue growth monitoring, backlog conversion updates, and AI chip sector intelligence — delivered to your inbox.
Cerebras Post-IPO Tracker → View Trading Day Analysis Get AI Chip AlertsRelated Research
Get the Full Reports — One-Time $19 Each
Post-IPO analysis, cap table, S-1, investment breakdown. Instant PDF download.
Or get all 375+ reports for $49/mo