Markets love a pivot story, even when the numbers underneath look like they were assembled by a stress test. That was the setup for Bitfarms on Tuesday: a $284.5 million net loss for 2025, a gross loss, and yet the stock still rose 6.6%. [1]
The immediate takeaway is simple. Investors appear to be looking past the mining business that produced the loss and pricing in the company's shift toward high-performance computing (HPC) and AI infrastructure. Sure, that does not make the 2025 results pretty. It does explain why the market reaction was not.
Register for free and get unlimited access to all articles.
The numbers that moved the story
Bitfarms reported $229 million in full-year revenue, up 72% year over year. On paper, that is strong top-line growth. In practice, it was not enough to cover the operating reality of the business. Cost of revenue came in at $248 million, which pushed the company into a gross loss before overhead and other line items even entered the frame. [1]
The bottom line deteriorated further as general and administrative expenses increased and the company took a hit from the revaluation of its crypto holdings. Bitfarms recorded a $50.5 million loss from the change in fair value of digital assets in 2025, compared with a $26 million gain in 2024. That swing matters. For Bitcoin$62,485.11miners, treasury exposure can flatter results in good tape and punish them when prices move the wrong way. This time it punished them. [1]
Management tied part of the damage to a decline in Bitcoin$62,485.11 prices during the year. That explanation is believable enough, though it is only part of the story. A miner with rising revenue still posted a gross loss, which points to a more basic issue: mining economics were under pressure even before accounting adjustments made things worse.
Equity markets do not always trade on reported earnings. Sometimes they trade on what investors think a company is trying to stop being. Bitfarms is now five months into its pivot from pure Bitcoin mining toward HPC and AI, and that appears to be the part of the update shareholders wanted to hear. [1]
The logic is familiar across the mining sector. Bitcoin mining is volatile, capital intensive, and increasingly margin-sensitive after the 2024 halving cut block rewards in half. HPC and AI hosting, by contrast, is being marketed as a steadier infrastructure business with potentially better contract visibility, assuming a company has the power access, data center footprint, and capital to execute. [2]
That last clause matters. Announcing an AI pivot is easy. Building a competitive HPC business is the expensive part, because of course it is. Investors still gave Bitfarms some credit for moving in that direction, likely seeing the shift as a way to monetize power assets beyond mining.
What the loss says about the core business
The 2025 results suggest Bitfarms' legacy model remains exposed on several fronts at once: Bitcoin$62,485.11 price volatility, high operating costs, and balance-sheet sensitivity to digital asset valuations. Those risks are not new, but the scale of the annual loss makes them harder to dismiss as temporary noise.
A 72% revenue increase would usually buy a company some breathing room. Here, it did not. With cost of revenue running above sales, Bitfarms effectively needed either better mining margins, stronger Bitcoin price support, or a faster contribution from its non-mining strategy. It had none of the three in sufficient size.
That is the deeper irony behind the stock move. Tuesday's rally was less a vote of confidence in the year that just ended and more a bet that the worst of the miner-only economics may not define the next one.
Bitfarms' strategy shift puts it in a growing list of crypto miners trying to repurpose energy access and data center capacity for AI-related workloads. In plain English, the pitch is this: if you already control power-rich industrial sites, maybe you can rent compute infrastructure to customers who need it for machine learning, cloud services, or other heavy processing tasks.
That can be attractive, but it is not a plug-and-play conversion. HPC tenants care about cooling design, uptime guarantees, networking, latency, and contract structure. Bitcoin miners care mostly about hash rate efficiency and power cost. There is some overlap, but not enough to pretend the transition is automatic.
So the market's positive read on Bitfarms likely reflects optionality, not proof. Investors are valuing the possibility that these assets can be repositioned into a more durable business line. Whether that turns into recurring cash flow is still an open question.
What to watch next
Three things matter from here.
First, gross margin. Revenue growth alone is not useful if the company still cannot cover direct costs. Any improvement there would tell investors the mining business is stabilizing or that new infrastructure is being deployed more efficiently.
Second, capital allocation around the HPC and AI buildout. Watch for specific site conversions, customer announcements, and funding requirements. Vague strategy language is cheap. Signed contracts are better.
Third, Bitcoin price exposure. Bitfarms is still meaningfully tied to crypto market moves, both operationally and through digital asset valuation. If Bitcoin weakens again, the company's legacy business could remain a drag while the newer segment is still ramping.
For now, the market seems willing to tolerate a very bad year in exchange for a potentially less cyclical future. That is a real thesis. It is also, at this stage, still a thesis.
Your reviews help us improve the quality of both current and future articles. All reviews are public and visible to other readers. We use both ratings and comments to improve future articles and to revise any articles that do not meet our standards.
See more like this on Google
