Fermi (FRMI) Q4 2025 Earnings Call Transcript

Fermi reported a $486 million full-year net loss, but most of it was noncash, with operating cash burn limited to $34 million and year-end cash of $409 million. The company has deployed $570 million into Project Matador, expects more than $3 billion of total Phase 0/1 capital deployment, and says liquidity is sufficient for at least 12 months. Management reiterated that no definitive tenant lease has been signed yet, while noting strong interest from hyperscalers and chipmakers and additional post-year-end financing facilities totaling $785 million.

Analysis

The market is still treating this as a binary “tenant or bust” story, but the more interesting edge is financing optionality versus execution drag. The near-term bull case is not revenue; it is that the company has effectively front-loaded the expensive, hard-to-permit, hard-to-source parts of the platform, which makes the next incremental lease disproportionately valuable if it converts into project debt. That should keep strategic value intact even if the first monetization step is delayed into 2027.

The second-order risk is that the current structure creates a mismatch between asset readiness and customer readiness. Management is signaling that power availability is ahead of MEP build-out, which means the bottleneck has shifted from power supply to tenant ecosystem coordination; that widens the set of counterparties but also lengthens diligence cycles and increases renegotiation risk. The implication is that headline demand can stay strong while transaction velocity remains slower than the market expects.

From a trading perspective, the lockup expiration is a near-term overhang because it introduces supply just as the stock is still a story stock with no operating revenue to anchor valuation. The more subtle issue is dilution of narrative quality: if the company has to rely on a broader set of smaller or more phased deals, the market may compress the implied value of the first gigawatt even if total long-run capacity stays attractive. On the other hand, the nonrecourse financing stack and REIT pathway create a cleaner asset-level leverage story than most pre-revenue infrastructure IPOs, which should limit downside if the lease process stalls rather than fails.

The contrarian view is that the market may be underestimating how much the permit and site completion de-risk the eventual lease-up, while overestimating how quickly that translates into signed contracts. If chipmakers are truly engaging more directly, the demand pool is broader than hyperscalers alone, but those conversations likely require tighter coordination around MEP timing and modular deployment. That favors patience and structure over outright directionality until a definitive lease is announced.

Key Decisions for Investors

• Avoid owning the common outright into the next 4-8 weeks unless you have event tolerance; lockup expiration plus no signed lease makes FRMI a poor risk/reward for cash equity at current stage.
• If expressing bullishness, prefer a defined-risk call spread in FRMI 3-6 months out rather than stock ownership; upside depends on one lease catalyst, while downside is driven by time decay and secondary supply.
• Pair trade: long MUFG / short FRMI into the financing milestone window. MUFG benefits from warehouse-finance visibility while FRMI remains execution-sensitive and supply-heavy; use FRMI as the leg only if borrow is available.
• For a hedge-fund style expression, buy FRMI only after a definitive tenant agreement and project-finance close; until then, the better trade is watching for a post-news pullback to add, not front-running management's optimism.
• If the market starts pricing FRMI like a funded infrastructure asset before leases are signed, fade strength via short-dated puts or put spreads; the key risk is multiple compression if lease timing slips by even one quarter.