Goldman Sachs Maintains Neutral Rating on CoreWeave with $85 Target Price

Overview

Goldman Sachs released a research report on CoreWeave (CRWV), maintaining its 'Neutral' rating and $85 target price (implying 34% downside). The report notes that although the company’s first-quarter revenue exceeded market expectations by 5%, its share price declined 9% after hours due to a $500 million upward revision in capital expenditures driven by rising component prices. Goldman Sachs believes CoreWeave’s unit economics remain robust, supported by its take-or-pay contract structure—existing transaction capacity is locked in, and pricing for new contracts can be adjusted to reflect higher input costs. Demand continues to accelerate, backlog has expanded significantly, and 2026 capacity is essentially sold out. Additionally, monetization of the software business is viewed as a mid-term profit enhancement opportunity.

Core views

Demand Acceleration and Backlog Growth: CoreWeave’s demand is accelerating, with revenue backlog reaching $99.4 billion, a substantial 49% increase quarter-over-quarter. Q1 marked the first quarter of the Vera Rubin transaction, while demand remains strong for Blackwell, Hopper, and Ampere chips. Average prices for A100, H100, H200, and L40 chips rose sequentially. This backlog reflects contracts currently online or expected to go live in 2026 and 2027. In fact, CoreWeave’s 2026 capacity is already nearly fully sold out. Its customer base is also diversifying, with commitments from non-investment-grade AI-native companies and foundational labs falling below 30% of total backlog. The company expanded its contract with Jane Street, signing an additional $6 billion in capacity during Q1; contracts with physical AI and spatial computing customers—including World Labs—exceed $1 billion in revenue backlog. Capacity Expansion and Capital Expenditures: Active power capacity has reached over 1 GW (previously 850 MW), and contracted power capacity has increased to 3.5 GW (previously 3.1 GW). Management reaffirmed its expectation to reach 1.7 GW of active power by end-2026 and now expects most of the 3.5 GW of contracted power to come online by end-2027 (previously guided at 3.1 GW), supporting a target of over $30 billion in annual recurring revenue (ARR) by end-2027—75% of which is already under contract. Capital expenditures were raised by $500 million at the midpoint, primarily reflecting higher component prices. Unit Economics and Software Opportunity: Goldman Sachs views CoreWeave’s unit economics as resilient, supported by take-or-pay contracts: capacity under existing transactions is locked in, and pricing for new contracts can be adjusted to offset higher input costs. CoreWeave prices consistently on contribution margin and retains confidence in its long-term operating margin target of 25–30%. Beyond GPU infrastructure, management highlighted demand for CPU, storage, networking, software solutions, and developer tools—all integrated into a single platform. Over 90% of reserved-instance customers use at least two products, and 75% use three or more. The storage business continues to grow (ARR of $100 million as of Q3 2025); management expects similar growth trajectories for software, CPU, and networking businesses, all projected to exceed $100 million ARR by end-2026. CoreWeave has also begun offering CoreWeave Omni, enabling customers to deploy CoreWeave’s cloud stack within their own data centers—a solution generating early interest among cloud, enterprise, and sovereign clients. Expansion of software offerings meaningfully broadens the addressable market and introduces higher-margin revenue streams, opportunities not included in current guidance and thus representing additional upside potential. Competitive Advantages and Risk Balance: Goldman Sachs continues to view CoreWeave’s architecture—purpose-built for high-end AI compute—as a key differentiator versus hyperscale cloud providers and other emerging cloud providers, especially in supply-constrained environments. However, this advantage is balanced against execution risk related to infrastructure deployment timelines, and limited visibility into how quickly CoreWeave can extend its offerings beyond high-density compute into higher-level services.

Analysis framework

Goldman Sachs’ analytical framework centers on 'unit economics' and 'demand/supply alignment.' First, by dissecting revenue, EBIT margin, and capital expenditure changes in the earnings report, it identifies rising capital expenditures—not deteriorating fundamentals—as the primary driver of share price volatility. Second, it uses 'backlog' and 'contracted power capacity' as leading indicators to validate the authenticity and sustainability of demand, particularly assessing capacity coverage for 2026–2027. Finally, it evaluates the monetization potential of non-core hardware revenue streams—such as software—to assess future profit uplift and valuation optionality, while weighing execution risks and competitive dynamics.

Methodology notes

• Industry/Vertical Analysis FrameworkSupply-demand framework

  Supply-Demand Framework

  The report analyzes the alignment between supply (capacity, power capacity) and demand (backlog, chip-type preferences) in the AI compute market to assess the company’s growth certainty and pricing power. In supply-constrained environments, firms with locked-in capacity hold a distinct advantage.

• Company Fundamentals and Financial Framework

  Unit Economics Analysis

  The report focuses closely on 'unit economics'—i.e., profitability per unit of compute or per contract. By analyzing the take-or-pay contract mechanism, it assesses whether the company can maintain stable contribution margins amid rising input costs through pricing adjustments on new contracts—thereby evaluating its resilience to cost shocks.

• Valuation MethodologyEV/EBITDA valuation

  EV/EBITDA Valuation

  The report lists EV/EBITDA multiples in key data tables as one of the core metrics for assessing valuation, especially appropriate during stages when the company has not yet achieved stable positive net income—making this multiple a more reliable indicator of operating cash generation and enterprise value.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• CoreWeave Inc. (CRWV)

  Direct beneficiary, AI compute infrastructure provider

  Strengths

  Architecture purpose-built for high-end AI compute, delivering differentiation in supply-constrained environments; take-or-pay contracts safeguarding unit economics; strong demand and near-full sell-out of 2026 capacity; software business providing high-margin incremental revenue.

  Weaknesses

  Capital expenditures sensitive to component price volatility; execution risk around infrastructure deployment timing; limited visibility into pace of expansion into higher-level services.

  Comparison

  Architectural advantage over hyperscale cloud providers and other emerging cloud providers in the high-end AI compute segment.

  Risks

  Reduction in overall GenAI compute demand; intensifying competition from hyperscale cloud providers and emerging cloud providers; disruption in data center equipment procurement; difficulty raising additional financing at competitive rates.

Key data

• Target Price$85.00Implies 34.0% downside
• Revenue Backlog$99.4BUp 49% quarter-over-quarter
• Capital Expenditure Adjustment+$500MMidpoint increase, reflecting higher component prices
• Contracted Power Capacity3.5 GWPreviously 3.1 GW; majority expected online by end-2027
• Active Power Capacity>1 GWPreviously 850 MW
• 2027 ARR Target>$30B75% already under contract
• Storage Business ARR$100MAs of Q3 2025

Impact & implications

The report concludes that CoreWeave’s share price reaction stems largely from concerns about rising capital expenditures—but these increases actually reflect higher component prices, not declining operational efficiency. Due to its unique contract structure, the company can pass through cost pressures to new customers, preserving unit economics stability. Longer term, software expansion and architectural differentiation in high-density computing will support its 25–30% operating margin target. For investors, focus should shift from short-term capex fluctuations toward progress on long-term capacity delivery and software monetization.

Risks

• Reduction in overall GenAI compute demand
• Intensifying competition from hyperscale cloud providers and emerging cloud providers
• Disruption in data center equipment procurement
• Difficulty raising additional financing at competitive rates
• Execution risk related to infrastructure deployment timelines
• Limited visibility into pace of expansion into higher-level services

What to watch

• Continuity of technological differentiation across future GPU generations
• Path to positive free cash flow—whether faster than expected
• Demand strength for CoreWeave’s broader software solutions
• Whether AI adoption continues to outpace supply growth