Alpha Compute Corp. Completes Wind-Down of Legacy Digital Asset Treasury, Returns TON Holdings

Alpha Compute (ALP) initiated the return of the final tranche of locked Toncoin/GRAM holdings to Animoca affiliates, removing ~$6 million of TON (GRAM) and the associated TON Put Option liabilities, and formally winding down its legacy Telegram Digital Asset Treasury. The company says it will hold only ~$200k of TON post-transaction, with future GRAM holdings limited to earned consideration for AI confidential compute delivered on Telegram’s Cocoon network. Management frames this as reducing mark-to-market balance-sheet volatility and aligning with a compute-first operating model; projected run-rate revenue is ~$23 million (unaudited) after the DAT wind-down.

Analysis

This is directionally positive for ALP because it removes a non-core source of volatility and should narrow the company’s valuation discount to a pure operating business rather than a quasi-treasury vehicle. That said, the market should not confuse balance-sheet cleanup with earnings quality: the economic value now depends almost entirely on utilization, pricing, and whether GPU leasing converts into durable gross profit, not on token markups.

The second-order shift is that token exposure has not disappeared; it has moved from the balance sheet into the revenue line. If GRAM is illiquid or volatile, ALP’s cash conversion may become lumpier even if reported revenue rises, which matters because lease liabilities remain large relative to forecast revenue. In the next 1-3 months, the key question is whether this de-risking helps lower the equity risk premium; over 6-18 months, the real catalyst is whether Cocoon becomes a real demand engine or just a strategic talking point.

The contrarian view is that the headline is small relative to ALP’s asset base and may be overread as a strategic breakthrough. The stock can re-rate on “cleaner story” optics in days, but if the next update does not show higher contracted utilization or improved cash flow, the market will likely fade the move and refocus on execution risk, financing needs, and customer concentration. Falsifiers: no ARR/throughput step-up over the next two quarters, rising GPU lease obligations, or any evidence that GRAM receipts are hard to monetize.

Key Decisions for Investors

• Trade the initial relief rally in ALP: if the stock gaps up materially on the announcement, use strength to reduce exposure or initiate a small 1-3 month put spread, because the headline is balance-sheet cosmetic unless operating metrics improve.
• Watch ALP for a real re-rating trigger: only add on evidence of higher GPU utilization, signed capacity expansion, or a step-up in quarterly revenue run-rate above the stated forecast; absent that, treat the name as a speculative execution story.
• Long-quality / short-story setup: prefer NVDA or a scaled GPU infrastructure beneficiary over ALP on a relative basis, since the market is likely to reward real infrastructure cash flow more than a microcap pivot narrative.
• Set a downside alert on GRAM liquidity and price stability; if the token weakens sharply or trading becomes thin, ALP’s new 'earned consideration' model could impair cash conversion and force a reset of the bullish thesis.
• No immediate trade in the token itself unless there is separate evidence that Cocoon adoption is accelerating; the return of the treasury tokens is too small to justify a standalone directional GRAM position.