What happened with AGPU today?
No read yet
Uses news and trading volume to explain the day’s move and how it relates to the market.
Stop guessing. Get the call, a three-month target price, the best and worst cases, and a source for every claim.
Axe Compute’s priority is building a business that supplies GPU-computing capacity through the decentralized Aethir network to enterprises and AI developers. Customers would reserve capacity under typically prepaid contracts, paying by GPU-hour rather than buying hardware or building data centers. The company does not own or operate physical data centers. Its compute and treasury segment has not yet generated reported revenue.
Alongside compute sales, Axe Compute holds ATH, Aethir’s token, which it can use to settle compute obligations, and plans to seek income through staking and other activities. This ties the business to Aethir’s network and a treasury concentrated in ATH, some of which is locked and cannot be freely transferred. Expanding compute capacity also depends on access to NVIDIA GPUs and supporting infrastructure; restrictions on exports of advanced chips to China constrain that supply chain. The company has relied on debt and equity funding because revenue has not covered its capital needs.
Its reported revenue comes entirely from drug-discovery services for biopharmaceutical clients. Its PEDAL platform uses tumor samples and historical data to predict drug responses, while laboratory services test drugs using proprietary tumor models. Axe Compute is seeking to sell this legacy business, operated through Helomics Holding Corporation, but a sale has not been completed.
The risk section is the longest, dullest part of the filing. This keeps only what can really hurt you: where it hits, how hard, when it plays out.
Every company claims an edge. Each one is checked against hard numbers like margins and returns: which hold, and which line gives way first.
Axe Compute’s outlook hinges on Aethir network adoption and ATH token prices: its priority compute business reported no revenue in 2025, while ATH revaluation drove substantial losses.
Alpha Compute competes with Axe Compute for enterprises and AI developers paying for GPU capacity to train and run AI models. Its contracted GPU lease rates and capacity utilization offer a comparison for Axe Compute’s reserved-capacity business.
NVIDIA supplies the GPUs Axe Compute primarily buys to provide paid computing capacity to enterprises and AI developers, so hardware availability and prices constrain Axe’s expansion. Watch GPU supply, pricing and delivery times: shortages or delays can hold up customer orders and reduce capacity.
Bit Digital’s WhiteFiber business competes with Axe Compute for AI and machine-learning developers paying for cloud GPUs to train and run models. WhiteFiber’s cloud GPU revenue is the relevant comparison; it also earns money hosting customers’ equipment in its own data centers.
SANGRIX and Axe Compute both have exposure to Aethir adoption: SANGRIX sells Aethir cloud-rendering units to institutional and retail customers and owns GPU servers commercialized by third parties, while Axe plans to sell enterprises GPU capacity sourced through Aethir. Watch customer purchases of SANGRIX’s Aethir units for evidence of demand for that network.
HIVE Digital Technologies competes with Axe Compute to sell GPU computing capacity to enterprises running AI workloads, but also has substantial mining operations. Its conversion of mining facilities into enterprise AI data centers is more relevant to Axe’s competitive position than payments from mining pools.
Bitdeer’s AI cloud services compete with Axe Compute’s planned paid GPU services, although Bitcoin mining remains a major part of Bitdeer’s business. Follow its AI infrastructure and cloud-services business separately from self-mining revenue when assessing competition for Axe.
TeraWulf and Axe Compute both depend on enterprise and hyperscale customers spending on AI computing: TeraWulf leases data-center infrastructure, while Axe sells GPU capacity. TeraWulf’s long-term hosting leases and delivery of contracted capacity show those customers’ infrastructure commitments, rather than demand for GPU-hours directly.
Core Scientific and Axe Compute both depend on demand for power-intensive AI computing: Core Scientific provides data-center space and infrastructure for customers’ equipment, while Axe sells enterprises access to GPU capacity. Watch Core Scientific’s development and repurposing of facilities for AI workloads, rather than treating its mining operations as a measure of Axe’s market.
TeraWulf Inc. earningsWULF
TeraWulf's report will show whether large customers keep committing to and receiving AI data-center capacity, an indirect signal for demand around Axe Compute's enterprise computing business.
Read TeraWulf's long-term hosting leases alongside the capacity it has delivered. Steady commitments with delivery keeping pace would show that large customers are still investing in AI infrastructure, supporting the case for a market around Axe Compute; fewer commitments or lagging delivery would weaken that signal. TeraWulf leases data-center infrastructure, not GPU-hours like Axe Compute sells, so this is evidence of infrastructure spending, not direct proof of GPU demand.
Axe Compute Inc. earningsThis company
Axe Compute sells GPU capacity to enterprises and AI developers, and this report will show whether that business is starting to bring in revenue rather than leaving reported results dominated by ATH price swings.
Read Compute Services revenue alongside cash used in operating activities. Axe Compute reported no revenue from this business in 2025 or 2024; revenue starting to appear while operating cash use holds steady or falls would show that GPU capacity is becoming a paid business. Little or no revenue with rising cash use would mean the business has yet to take hold and still depends on financing or ATH assets. Separate changes in ATH fair value from that picture: they enter reported earnings but do not show whether the GPU business sold more or less.
Bit Digital, Inc. earningsBTBT
Bit Digital's WhiteFiber cloud GPU revenue will show whether AI developers are paying for cloud GPU capacity, the same kind of customer Axe Compute is trying to win.
Separate WhiteFiber's cloud GPU revenue from its revenue for hosting customers' equipment. Rising cloud GPU revenue would show developers are paying for cloud capacity and Bit Digital is winning those customers: that validates paid demand in Axe Compute's market but also makes the competition more tangible. If equipment hosting holds up while cloud GPU revenue stays weak, data-center activity is not translating into GPU cloud sales, and Axe Compute's own revenue remains the test of its demand.
Nvidia Corp earningsNVDA
NVIDIA is a primary GPU supplier to Axe Compute, so its report on demand, supply and delivery can show whether Axe Compute can add capacity in time to serve customers.
Read NVIDIA's description of GPU demand alongside what it says about supply, pricing and delivery times. Strong demand with more supply and improving delivery would suggest that a hardware constraint on Axe Compute's expansion may be easing; strong demand with limited supply, slower delivery or higher prices would make it harder for Axe Compute to obtain equipment on time and serve orders. Weaker demand might ease procurement pressure, but it would not show that Axe Compute's customers want more capacity.
Alpha Compute Corp earningsALP
Alpha Compute's GPU lease rates and capacity utilization will show whether customers are filling paid capacity at steady prices, a direct read on the pricing environment for Axe Compute.
Pair Alpha Compute's GPU lease rates with its capacity utilization. Firm rates and high utilization would show paid demand keeping up with available capacity, supporting prices in Axe Compute's market while showing that a rival can attract customers. Falling rates with low utilization would point to demand lagging capacity or competitors cutting prices, making it harder for Axe Compute to hold its rates. Together, the two measures help distinguish strong demand from excess capacity.
