Soluna Finishes Kati 1 and Secures Conditional Base Load Position for All 166 MW in ERCOT

Soluna Holdings completed energization of the final 14 MW at Project Kati 1 in Texas, bringing the green data-center site's total capacity to 83 MW. ERCOT also conditionally classified the full 166 MW Kati campus, including the planned Kati 2 AI expansion, supporting additional development capacity for AI computing and Bitcoin mining.

Analysis

SLNH’s equity value remains far more sensitive to conversion of contracted/monetized power capacity than to physical commissioning milestones. The strategic optionality is legitimate: dispatchable, renewables-adjacent load can command a premium versus conventional data-center development when grid congestion and AI power demand tighten. But conditional interconnection treatment does not establish customer commitments, power economics, construction funding, or returns on invested capital; absent these, the market is likely to discount the expansion as capital-intensive option value rather than recurring EBITDA.

Near term (days to 1 month), the announcement can support retail and thematic-AI/crypto flows in a thinly traded small-cap, but the move is vulnerable to reversal without disclosed counterparties, contract duration, pricing structure, and incremental financing terms. Over 1-3 months, the relevant catalyst is a binding AI hosting or HPC lease that validates revenue per MW materially above Bitcoin-mining economics. The key adverse scenario is dilution or expensive project debt before contracted cash flow, particularly if Bitcoin economics weaken and AI customers prioritize larger, investment-grade operators with proven delivery records.

The non-obvious competitive implication is that flexible-load projects may gain value from ERCOT volatility, but only if SLNH retains sufficient share of curtailment/grid-service economics rather than passing it through to customers. Larger private data-center developers and hyperscale-linked operators have lower cost of capital, so SLNH’s potential advantage is speed and stranded-power access—not balance-sheet capacity. A sustained valuation rerating requires evidence that its realized EBITDA per energized MW and uptime exceed mining-hosting alternatives after curtailment; headline MW alone is not a sufficient benchmark.

Key Decisions for Investors

• No immediate core position in SLNH; treat as an event-driven watch item until management discloses a signed customer, contracted MW, term, revenue/MW, expected capex, and funding source. The missing unit-economics disclosure is the gating item for underwriting.
• If SLNH announces a creditworthy AI/HPC customer with multi-year contracted revenue and non-dilutive funding, consider a small long initiated after liquidity normalizes, targeting a 3-6 month rerating from speculative capacity value to contracted EBITDA value. Size for high volatility and use a hard thesis stop on an equity raise at a material discount or reduced deployment guidance.
• For broader AI-power exposure, prefer liquid, financed infrastructure proxies such as EQT or VRT over SLNH until contract economics are verified; this captures data-center power and equipment demand with materially lower single-asset execution risk.
• Monitor Bitcoin hashprice and ERCOT curtailment data over the next quarter. A sharp hashprice decline, elevated curtailment without compensating grid-service revenue, or any delay in converting available capacity into contracted load would falsify the near-term SLNH upside thesis.