Decentralized physical infrastructure networks (DePIN) generated an estimated $150 million in on-chain revenue in January 2026 alone, according to BlockEden.xyz analysis — a figure that dwarfs the sector's estimated $72 million in total on-chain revenue for all of 2025, per Messari. The combined ...
"When token prices are flat, the only thing that matters is whether someone is paying for the service, and whether the network can sustain itself without subsidies. That shift is healthy." — Markus Levin, Co-founder, XYO
Decentralized physical infrastructure networks (DePIN) generated an estimated $150 million in on-chain revenue in January 2026 alone, according to BlockEden.xyz analysis — a figure that dwarfs the sector's estimated $72 million in total on-chain revenue for all of 2025, per Messari. The combined DePIN market capitalization sits near $10 billion across 440-plus tracked projects, with over 41 million physical devices deployed worldwide.
The data describes a sector at an inflection point. Revenue is rising while token prices remain 94–99% below all-time highs. Enterprise customers — not token speculators — now drive the majority of demand growth in compute, wireless, and storage verticals. Leading networks trade at 10–25x revenue multiples, compressed from 1,000x-plus during the 2021 cycle. The transition from token-subsidy dependence to fee-based sustainability is underway, but unevenly distributed. A handful of projects generate meaningful revenue; the long tail remains subsidy-dependent.
This report examines the revenue data, the structural economics driving the shift, and the risks that persist as DePIN attempts to prove it can function as infrastructure rather than speculation.
DePINscan, the sector's primary analytics platform, tracks 440 active projects across compute, bandwidth, energy, storage, wireless, and sensor verticals. These networks collectively support 40.9 million contributing devices — up from fewer than 10 million in mid-2023, representing approximately 4x growth in under three years.
The sector's combined market capitalization stands at approximately $6.5–10 billion, depending on classification methodology. Messari's narrower framework places the figure near $10 billion in circulating value. DePINscan's broader tracker reports $6.46 billion. The discrepancy reflects ongoing definitional ambiguity — whether AI-adjacent compute tokens, sensor networks, and energy grid projects all belong under the same umbrella remains unsettled.
DePIN startups raised approximately $1 billion in venture funding during 2025, concentrated at seed and Series A stages, according to Messari. This continued private-market investment, even as public token valuations remained depressed, suggests institutional conviction in the thesis that decentralized infrastructure can capture real demand.
The revenue picture is concentrated. A small group of projects generates the majority of verifiable on-chain income.
| Project | Vertical | Revenue Metric | Period | |---------|----------|---------------|--------| | Aethir | GPU Compute | $127.8M revenue; $147M ARR | FY 2025 / Q3 2025 | | Helium | Wireless | $11M annualized (excl. discretionary burn) | Q4 2025 | | Akash Network | Cloud Compute | $4.2M ARR | Q1 2026 | | io.net | GPU Compute | $20M+ cumulative since launch | Through 2026 | | Render Network | GPU Rendering | $300K peak weekly | 2025–2026 | | Hivemapper | Mapping/Sensors | ~$600K cumulative since Jan 2025 | Through Jan 2026 | | Filecoin | Storage | 6,500+ Filecoin Pay transactions | Through Nov 2025 |
Aethir dominates the revenue table by a wide margin. Its $127.8 million in 2025 full-year revenue — with 22% quarter-over-quarter growth through Q3 — is largely enterprise-sourced. The company reports 150-plus clients and partners, 1.5 billion compute hours delivered, and 440,000 GPU containers operating across 94 countries.
Helium's revenue, while smaller in absolute terms, demonstrates a different model. Average daily data credit burns grew 83.6% quarter-over-quarter to $56,760 in Q4 2025, with 43.9% attributable to mobile carrier offloading. The network offloaded 4,388 TB of carrier data in Q4 2025, up 60.7% from 2,731 TB in Q3. Total active subscribers surpassed 120,000 by February 2026.
At the other end of the spectrum, Hivemapper's $600,000 in cumulative revenue over twelve months and Filecoin's ambiguous storage revenue figures illustrate the gap between the sector's leaders and its middle tier.
The original DePIN economic model relied on a straightforward but fragile mechanism: networks issued inflationary tokens to incentivize hardware providers, bootstrapping supply ahead of demand. When token prices held, the model worked. When they fell, it triggered what researchers at Frontiers in Blockchain (2025) formalized as a "death spiral" dynamic: falling token prices reduce provider income, providers exit, network quality degrades, users leave, demand drops, and the token falls further.
As a 2025 Frontiers scoping review of DePIN tokenomics noted, "success hinges on transitioning from models subsidized by token inflation towards systems where real economic value, generated through user fees, sustains the network." The review, covering 24 academic papers and 22 grey-literature sources published between 2021 and mid-2025, identified governance concentration ("whale dominance"), regulatory ambiguity around token classification, and subsidy dependence as the sector's three structural risks.
The 2026 correction appears to be forcing this transition. With DePIN tokens down 94–99% from cycle highs, according to Messari, the subsidy model is mechanically broken for most networks — tokens simply are not worth enough to fund provider payouts at scale. What remains is real demand or nothing.
Several projects are explicitly engineering the transition. io.net's Incentive Dynamic Engine (IDE), expected Q2 2026, replaces token-denominated provider payouts with USD-stabilized compensation, dynamically adjusting token supply against real-time revenue. Akash Network's active lease count surged from approximately 400–500 to 800–1,000 in a single week in mid-March 2026, providing on-chain evidence of organic demand growth. Helium suspended its discretionary subscriber revenue burn program in January 2026 to focus on carrier offload revenue — the service that generates enterprise-grade demand rather than retail token burns.
The single largest demand catalyst for DePIN compute networks in 2026 is AI infrastructure overflow. US technology companies alone are projected to invest approximately $650 billion in AI infrastructure this year, with $450 billion allocated to GPU compute and data center buildout. Hyperscaler GPU capacity is genuinely constrained; wait times for NVIDIA H100 and B200 allocations remain measured in months, not weeks.
Decentralized compute networks offer 60–70% cost savings relative to AWS, GCP, and Azure pricing, according to multiple operator reports. Akash Network prices GPU compute at $1.20–$1.80 per hour versus $4.50–$5.50 per hour on AWS. Enterprise AI teams are migrating inference workloads — the compute-intensive but latency-tolerant tasks that suit distributed architecture — to decentralized networks not because of ideological alignment with decentralization but because the economics are compelling.
Aethir's client base spans gaming studios, AI inference providers, model training operations, and AI agent platforms. Render Network integrated NVIDIA's Blackwell architecture in Q1 2026 and onboarded approximately 60,000 GPUs from Salad Network through governance proposal RNP-023 in April 2026. AI workloads now represent 35–40% of Render's total job volume, up from a negligible share two years prior.
Volkswagen's autonomous-driving subsidiary ADMT signed with Hivemapper to source real-time road-mapping data from 100,000-plus contributor dashcams for its 2026 driverless fleet operations. This represents a different flavor of enterprise demand — not compute, but verifiable real-world data — but the economics follow the same pattern: a centralized buyer paying for a decentralized service because the price and coverage outperform traditional procurement.
Compute is the largest revenue vertical by a significant margin. Aethir, Akash, io.net, and Render collectively account for the majority of verifiable DePIN revenue. The AI infrastructure shortage provides a structural demand floor that did not exist during the 2021 cycle. The risk is concentration: if hyperscaler capacity constraints ease — NVIDIA's supply chain is expected to normalize through late 2026 — the cost arbitrage narrows.
Wireless is the most mature vertical in terms of real-world deployment. Helium's 120,000-plus mobile subscribers, 2.5 million peak daily active users (December 2025), and carrier partnerships with AT&T, Movistar, and Free Mobile represent genuine infrastructure adoption. Carrier offloading — where traditional telcos route data through Helium hotspots to reduce backhaul costs — is an established telecom procurement model, which reduces adoption friction. The revenue remains modest ($11 million annualized) relative to the network's operational scale.
Storage is the most complex vertical to evaluate. Filecoin, the largest decentralized storage network, has shifted its 2026 strategy toward "aligning incentives with paid usage," an implicit acknowledgment that historical storage deals were overwhelmingly subsidy-driven. The Filecoin OnChain Cloud (FOC) upgrade launched in November 2025 to simplify commercial onboarding. Institutional clients include the Internet Archive, MIT Open Learning, and the Smithsonian. Whether the network can convert its massive raw capacity into paid enterprise revenue remains the sector's largest open question.
Sensors and mapping remain early-stage by revenue metrics. Hivemapper has mapped approximately 33% of the global road network (644 million km total, 21 million km unique) using over 100,000 dashcams, and its shift from $589 upfront hardware costs to $19/month subscriptions aims to accelerate contributor growth. The Volkswagen partnership validates the data's quality, but $600,000 in cumulative revenue over twelve months suggests the demand side has not yet scaled to match supply.
The divergence between DePIN token prices and DePIN network revenues is the sector's defining feature in 2026. As Messari senior analyst Dylan Bane noted, "DePINs must prioritize finding product-market fit on the demand side." Token markets appear to be pricing exactly this uncertainty.
DePIN tokens launched between 2018 and 2022 trade at 94–99% discounts to all-time highs. Meanwhile, leading revenue-generating networks carry valuation multiples of 10–25x revenue — a dramatic compression from the 1,000x-plus multiples of the 2021 cycle. At these levels, DePIN projects can be evaluated using traditional infrastructure metrics: revenue growth, customer acquisition cost, unit economics, and retention.
This compression has a silver lining. DePIN can now attract capital that evaluates infrastructure on cash-flow fundamentals rather than narrative momentum. The emerging "InfraFi" concept — leveraging the $175-billion-plus stablecoin market to provide fiat-denominated infrastructure financing — aims to bridge the gap between crypto-native token models and institutional capital allocation frameworks.
Demand concentration. AI compute overflow is the primary growth driver. If hyperscaler capacity normalizes — and NVIDIA, AMD, and Intel are all expanding production — the cost advantage narrows. DePIN compute networks must develop sticky enterprise relationships before the supply constraint lifts.
Regulatory ambiguity. The Frontiers scoping review identified "regulatory fog" as a structural risk. Token classification (security vs. utility), infrastructure-provider licensing, and data-privacy compliance across jurisdictions remain unresolved. No major regulatory framework specifically addresses decentralized infrastructure networks.
Governance fragility. Low voter turnout and whale dominance in on-chain governance can stall critical tokenomic updates. The Frontiers paper noted these dynamics can "freeze protocol evolution or concentrate control in the hands of a few."
Revenue quality. Not all reported revenue is equivalent. Token burns driven by discretionary spending (e.g., Helium's suspended subscriber burn program) differ materially from enterprise contract revenue (e.g., Aethir's client agreements). Investors should scrutinize whether reported figures reflect genuine customer payments or internal token mechanics.
Long-tail fragility. Of 440-plus tracked projects, fewer than ten generate material revenue. The remainder operate on token subsidies that are mechanically unsustainable at current price levels. A consolidation event appears probable.
DePIN in May 2026 presents a sector with a clear thesis and uneven execution. The data supports two simultaneous conclusions that are not contradictory: real enterprise demand exists and is growing, and the majority of projects have not yet proven they can capture it.
The January 2026 revenue figures — $150 million in monthly on-chain payments — represent a genuine milestone. But context matters. Aethir alone likely contributed the majority of that figure. Strip out the top five projects and the remaining 435 networks collectively generate revenue that would be immaterial on a traditional infrastructure company's income statement.
The economic question is whether DePIN can sustain fee-based revenue growth as AI infrastructure supply constraints ease. The networks that survive will be those that converted their current cost advantage into durable enterprise relationships before the window closes. The rest face the same death spiral the original tokenomics papers warned about — only without the token price tailwind to delay it.
For investors evaluating DePIN, the relevant metric is no longer market capitalization or device count. It is recurring revenue from non-crypto-native customers, denominated in dollars, verified on-chain.