The decentralized physical infrastructure network (DePIN) sector has entered a defining phase. After a brutal 2025 that saw DePIN tokens decline 76.7% as a category — among the worst-performing segments in crypto — the sector's underlying economics tell a radically different story. On-chain reven...
"DePIN is being forced into fundamentals. When token prices are flat, the only thing that matters is whether someone is actually paying for the service, and whether the network can sustain itself without subsidies." — Markus Levin, Co-founder, XYO
The decentralized physical infrastructure network (DePIN) sector has entered a defining phase. After a brutal 2025 that saw DePIN tokens decline 76.7% as a category — among the worst-performing segments in crypto — the sector's underlying economics tell a radically different story. On-chain revenues hit $72 million in FY2025, Solana-based DePIN protocols posted an all-time monthly revenue high of $2.6 million in January 2026, and private investors poured approximately $1 billion into DePIN startups at seed and Series A stages throughout 2025.
This divergence between collapsing token prices and rising real revenues is not a bug — it is the market performing a violent repricing from speculation to fundamentals. Leading DePIN networks now trade at 10–25x revenue, down from over 1,000x in the 2021 cycle. For institutional investors, this compression represents either a death spiral or a generational entry point. The data increasingly supports the latter.
The DePIN sector has grown to roughly $19 billion in total market capitalization, spanning 13+ million active devices across compute, wireless, mapping, positioning, and sensor networks. Its 2026 trajectory is being shaped by three converging forces: insatiable AI compute demand, enterprise telecom adoption, and the emergence of real unit economics that can survive without token subsidies.
The DePIN sector's 2025 performance created one of the most dramatic disconnects in crypto history. While the combined market capitalization of DePIN tokens fell by more than three-quarters, aggregate on-chain revenue grew consistently. Messari's January 2026 "State of DePIN" report documented $72 million in verifiable on-chain revenue for the sector in 2025 — a figure that, while modest by traditional infrastructure standards, represents real payments from real customers for real services.
The valuation compression has been severe. Projects launched between 2018 and 2022 now trade 94–99% below their all-time highs. But this compression has also reset the valuation framework. Leading revenue-generating networks now trade at 10–25x revenue, placing them in territory comparable to early-stage SaaS companies rather than speculative crypto assets.
This repricing is forcing a Darwinian selection. Networks without genuine revenue streams are being starved of capital and attention, while those demonstrating real demand — Aethir, Helium, Render, GEODNET — are attracting disproportionate private investment. Messari identifies DePIN as the only crypto sector it expects to generate hundreds of millions of dollars in verifiable revenue by 2026.
The AI infrastructure arms race is DePIN's strongest tailwind. U.S. tech companies alone are projected to spend approximately $650 billion on AI infrastructure by 2026, according to Bridgewater Associates. Yet semiconductor supply remains constrained — SK Hynix and Micron have announced their entire 2026 output is pre-sold. This supply-demand imbalance creates a structural opening for decentralized compute.
Aethir has emerged as the sector's revenue leader. The decentralized GPU cloud platform generated $127.8 million in revenue across 2025 and reached $166 million in annualized recurring revenue (ARR) by Q3, serving 150+ enterprise clients across AI training, inference, cloud gaming, and real-time rendering. Its network spans 435,000 GPUs across 93 countries. These are not speculative token flows — this is enterprise compute spending, routed through decentralized infrastructure. In Q1 2026, Aethir aims to more than double its global compute footprint through its Strategic Compute Reserve, which onboards institutional AI clients.
Render Network reported a 40% increase in network compute power in 2025, rendering 63 million cumulative frames, with 35% of output attributed to Hollywood studios and AI training clients. At CES 2026, the team showcased partnerships with NVIDIA, Stability AI, and Luma Labs aimed at meeting explosive demand for edge ML workloads. Render's Burn-Mint Equilibrium (BME) token model has already burned over 1 million RENDER tokens through actual compute jobs.
Akash Network has carved a distinct niche with its reverse-auction model, where GPU providers compete for workloads. GPU utilization holds near 80%, and the Starcluster initiative — involving the acquisition of approximately 7,200 NVIDIA GB200 GPUs operated by vetted enterprise-grade datacenter "Nodekeepers" — positions the network to serve hyperscale AI demand as hardware comes online through early 2026.
The critical insight: inference — the running of trained AI models — accounts for roughly 80% of all GPU workloads. Unlike training, which requires tightly coupled supercomputer clusters, inference is embarrassingly parallelizable and geographically distributable. This is precisely the workload profile that favors decentralized architecture.
Helium's evolution from a hobbyist hotspot project to a legitimate telecom infrastructure provider represents DePIN's most compelling enterprise adoption case.
In Q4 2025, Helium's network offloaded 4,388 terabytes of data from major U.S. mobile carriers — a 60.7% increase from Q3's 2,731 TB. The network surpassed 2 million daily active users for the first time in Q4, peaking at 2.5 million on December 20, 2025. By February 2026, Helium Mobile had crossed 120,000 active subscribers.
Helium's monthly revenue exceeded $2.2 million by January 2026, contributing to the Solana DePIN ecosystem's all-time monthly revenue record of $2.6 million. Revenue is now evenly split between subscriber fees and carrier data offloading — a diversification that reduces dependence on retail consumer acquisition.
The partnerships tell the broader story. Helium's integrations with AT&T, Telefónica, Volkswagen, and DISH Network signal that decentralized wireless is being treated as legitimate complementary infrastructure by Fortune 500 companies. The blockchain market in telecom is projected to grow from $1.07 billion in 2024 to $7.25 billion by 2030.
XNET delivered 41x revenue growth from December 2025 to January 2026, and alongside Helium, posted over 600% revenue growth year-over-year from January 2025 — the strongest performance across the entire DePIN landscape.
Below the headline compute and wireless networks, a tier of specialized DePIN projects is demonstrating that niche infrastructure services can generate sustainable economics.
GEODNET — the world's largest decentralized real-time kinematic (RTK) positioning network — operates 21,000+ active stations across 145 countries. Its Q3 2025 revenue reached $1.23 million (216% year-over-year growth), with annualized revenue climbing from $5 million toward $7.3 million. VanEck has published a public bull case on GEODNET, reflecting growing institutional interest in precision positioning infrastructure for autonomous vehicles, agriculture, and construction.
Hivemapper has mapped over 700 million kilometers of roads — approximately 37% of the world's road infrastructure — through its decentralized dashcam network. Its mapping subsidiary Bee Maps secured $32 million led by Pantera Capital to scale AI-powered mapping. While quarterly revenue remains modest at $138,000 in Q4 2025, the underlying data asset — continuously updated, crowd-sourced global road maps — serves a market dominated by expensive, slow-updating incumbents.
These projects exemplify a DePIN design pattern: coordinating distributed physical sensors to produce data products that compete with centralized incumbents on freshness, coverage, and cost.
The economic value framework for DePIN departs fundamentally from traditional crypto tokenomics. Rather than value accruing purely through token speculation or MEV extraction, DePIN networks generate revenue through the sale of actual infrastructure services — compute cycles, wireless connectivity, mapping data, positioning signals.
The value distribution chain in a mature DePIN network flows through several participants:
The critical question is whether DePIN's unit economics can survive without token subsidies. The evidence is mixed but trending positive. Aethir's $166 million ARR from 150+ enterprise clients represents genuine commercial demand. Helium's 50/50 revenue split between subscribers and data offloading suggests a dual-revenue model that could sustain operations independently. GEODNET's 216% year-over-year revenue growth demonstrates that niche infrastructure with real buyers can scale.
However, many DePIN networks still rely heavily on token emissions to subsidize node operators — a model that compresses as token prices fall. The networks that transition to self-sustaining revenue models will survive; those that cannot will join the 94–99% drawdown as a permanent condition.
Regulatory ambiguity. DePIN networks that sell telecommunications or compute services operate in heavily regulated sectors. Helium's carrier partnerships work within existing frameworks, but decentralized wireless networks that bypass licensing requirements face potential enforcement action as they scale.
Centralization pressure. As enterprise clients demand SLAs, uptime guarantees, and compliance, DePIN networks face pressure to centralize quality control. Akash's "Nodekeepers" and Aethir's Strategic Compute Reserve represent moves toward managed infrastructure that may compromise the decentralization thesis.
Token overhang. Despite revenue growth, the massive gap between current token prices and historical highs creates a persistent overhang of potential sellers. Networks must demonstrate that revenue growth can outpace sell pressure from early investors and team unlocks.
Competition from hyperscalers. AWS, Azure, and GCP are investing aggressively in edge compute and regional deployments. DePIN's cost advantage narrows as hyperscalers improve pricing and geographic coverage.
DePIN tokens fell 76.7% in 2025, but on-chain revenues grew to $72 million — the most extreme token-price-to-revenue divergence in crypto, creating compressed valuations of 10–25x revenue.
Aethir's $166 million ARR from 150+ enterprise clients proves that decentralized GPU infrastructure can compete for real enterprise compute budgets, not just crypto-native demand.
Helium crossed 120,000 subscribers and 2 million daily active users, with data offloading to major carriers growing 60.7% quarter-over-quarter — validating decentralized wireless as complementary telecom infrastructure.
Solana-based DePIN protocols hit an all-time monthly revenue record of $2.6 million in January 2026, with Helium and XNET leading at 600%+ year-over-year growth.
Private capital remains aggressive: approximately $1 billion invested in DePIN startups in 2025 at seed and Series A, even as public token markets priced in doubt.
AI inference workloads — 80% of GPU demand — are structurally suited to decentralized networks, creating a secular tailwind as compute demand outstrips centralized supply.
DePIN is undergoing the transition that separates infrastructure from speculation. The 2025 token crash was not a failure of the thesis — it was a repricing that eliminated leveraged speculation and forced networks to prove economic viability through actual revenue. The networks that emerged — Aethir, Helium, Render, GEODNET, Akash — are generating real cash flows from real customers, building partnerships with Fortune 500 enterprises, and attracting institutional capital.
The sector's trajectory now hinges on a single question: can decentralized infrastructure networks achieve the unit economics, reliability, and compliance standards that enterprise buyers require? The early evidence — $166 million in ARR at Aethir, 120,000 mobile subscribers at Helium, enterprise partnerships at Render — suggests the answer is increasingly yes.
For investors and allocators, DePIN at 10–25x revenue multiples represents a fundamentally different proposition than DePIN at 1,000x. The speculative premium has been destroyed. What remains is infrastructure that either works or doesn't. That clarity, painful as it has been for token holders, may be exactly what the sector needed.