Decentralized Physical Infrastructure Networks generated an estimated $150 million in on-chain revenue in January 2026 alone, according to BlockEden data — a figure that dwarfs the $72 million the entire sector produced across all of 2025, per Messari's "State of DePIN 2025" report. The revenue a...
"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 Network
Decentralized Physical Infrastructure Networks generated an estimated $150 million in on-chain revenue in January 2026 alone, according to BlockEden data — a figure that dwarfs the $72 million the entire sector produced across all of 2025, per Messari's "State of DePIN 2025" report. The revenue acceleration is concentrated in GPU compute and bandwidth networks serving enterprise AI workloads, with Aethir ($166 million ARR), Render ($38 million in January 2026 revenue), and Grass ($33 million ARR) accounting for the bulk of commercial activity.
The paradox: tokens are not following. Legacy DePIN tokens launched between 2018 and 2022 remain 94–99% below all-time highs, according to Messari. The sector's combined market capitalization sits at roughly $8.9–10 billion — a fraction of the $19 billion peak reached in September 2025. Valuation multiples have compressed from 1,000x+ revenue during the 2021 cycle to 10–25x today. What remains is a sector that increasingly resembles a traditional infrastructure business: capital-intensive, margin-sensitive, and valued on cash flow rather than narrative.
The scale of the revenue shift is difficult to overstate. Messari's January 2026 report estimated the entire DePIN sector generated $72 million in on-chain revenue throughout 2025. By January 2026, leading networks were generating approximately $150 million in a single month — an 800% year-over-year jump for top performers, according to BlockEden analysis.
This is not evenly distributed. A small number of networks — primarily those offering GPU compute, AI training data, and decentralized wireless — account for the vast majority of commercial revenue. The long tail of the sector's 650+ tracked projects, per DePINscan data from March 2026, remains subsidy-dependent or pre-revenue.
The revenue composition has also changed. In prior cycles, on-chain revenue largely consisted of token-incentivized activity: users farming rewards, not purchasing services. The current cohort of revenue leaders is generating income from enterprise customers paying for compute cycles, storage deals, bandwidth, and verified data — services priced in fiat or stablecoins, settled on-chain.
Aethir — the sector's largest revenue generator — reported $147 million in annualized recurring revenue by Q3 2025, with quarterly growth accelerating from 14.5% (Q1 to Q2) to 22% (Q2 to Q3). The network operates 440,000+ GPU containers across 94 countries and serves 150+ enterprise clients. In April 2026, Axe Compute (NASDAQ: AGPU) closed a $260 million multi-year enterprise contract centered on a 2,304-GPU NVIDIA B300 cluster powered by Aethir's infrastructure — one of the largest single deals in decentralized cloud history.
Render Network generated $38 million in revenue in January 2026 alone, driven by AI inference and 3D rendering workloads. The network has burned 121 million RNDR tokens through its burn-and-mint mechanism, and at peak processed 1.49 million frames in a single month (July 2025). Render shipped Octane 2026 in early Q1 as NVIDIA flagged surging AI compute demand.
Grass monetizes unused internet bandwidth from 8.5 million users, providing verified web data to AI training pipelines. Revenue grew from approximately $2.75 million in Q2 2025 to $4.3 million in Q3 and a projected $12.8 million in Q4. The network's $33 million ARR is underpinned by enterprise contracts with multiple major AI organizations, including deals for white-labeled video downloading infrastructure. Grass raised a $10 million bridge round to fund expansion.
Helium hit a $22.4 million annualized run rate by December 2025, with average daily data credit burns rising to $56,760 — a 196.6% quarter-over-quarter increase by Q3 2025. The network's 541,000+ subscribers use T-Mobile and AT&T infrastructure as roaming partners across 115,000+ deployed hotspots. Grayscale added HNT to its Q2 2026 "Assets Under Consideration" list. Notably, Helium's CEO suspended a 100% revenue-to-burn experiment in January 2026 to prioritize user growth and carrier offload deals.
Akash Network reached $4.2 million in ARR, with Q1 2025 lease revenue of $1 million (38% QoQ increase). Provider count grew approximately 40% year-over-year to 120+ active providers by March 2026. Akash's pricing — H100 GPUs at $1.20–$1.80/hour versus AWS at $4.50–$5.50/hour — represents 60–70% cost savings that enterprise buyers are beginning to act on.
Filecoin is pivoting from capacity accumulation to revenue conversion. The network operates at exbibyte scale but generated only $180,700 in protocol fees over a recent 30-day period. Filecoin launched its Onchain Cloud platform in November 2025, attracting 100+ building teams and processing 6,500+ payment transactions through Filecoin Pay. The community has identified AI agents and enterprise data pipelines as the primary go-to-market verticals for 2026.
DePIN's most striking feature in 2026 is the divergence between operational performance and token valuation. According to Messari senior research analyst Dylan Bane, there are "no obvious catalysts to increase investment" in the sector despite improving fundamentals.
The numbers are stark:
| Metric | 2021 Cycle | May 2026 | |--------|-----------|----------| | Revenue-to-valuation multiple | 1,000x+ | 10–25x | | Legacy token drawdown from ATH | — | 94–99% | | Sector market cap | ~$19B peak (Sept 2025) | ~$8.9B | | On-chain revenue (annualized) | Negligible real revenue | $150M/month run rate |
The compression reflects a broader market recalibration. In favorable conditions, Bane noted, "gimmicks" such as ecosystem partnerships and community incentives "can actually help accelerate the build out of supply side growth." In the current environment, only unit economics matter.
This creates an unusual investment dynamic: DePIN networks are generating more real revenue than at any point in crypto history, yet their tokens trade at a fraction of prior valuations. For the sector, the question is whether sustained revenue growth eventually re-rates tokens — or whether the token model itself is structurally impaired for infrastructure businesses.
The revenue inflection is not occurring in a vacuum. Global enterprise AI infrastructure spending is projected to reach $650 billion in 2026, with $450 billion earmarked for GPU compute and data centers, according to industry estimates compiled by BlockEden. Global data center power demand is projected to reach 96 GW in 2026 (up from 49 GW in 2023), with AI workloads consuming approximately 40 GW.
Centralized hyperscalers — AWS, Azure, Google Cloud — cannot build capacity fast enough to meet demand. DePIN networks are absorbing overflow. Akash's Envision Labs case study documented a 30% reduction in GPU costs. Aethir delivers enterprise compute at approximately 70% below AWS equivalent pricing. The average cost savings across leading DePIN compute networks ranges from 50–85% versus centralized providers.
GPU utilization rates tell a complementary story: DePIN networks report 95%+ utilization versus 15–30% in centralized data centers, according to BlockEden. The economic logic is straightforward — aggregating idle GPU capacity from distributed sources eliminates the capital expenditure of building new data centers while achieving higher asset utilization.
The risk is concentration. AI compute demand is the primary growth engine. If AI spending slows, or if hyperscalers close the capacity gap, DePIN's cost advantage narrows. The sector's revenue base remains narrow and cyclically exposed.
With $175+ billion in stablecoins outstanding, a new financing model — termed "InfraFi" — is emerging as an alternative to venture capital and token sales for funding physical infrastructure deployment.
The mechanism: investors deposit stablecoins into protocol vaults. Vault capital finances the purchase and deployment of physical infrastructure assets (GPUs, hotspots, sensors, storage nodes). Revenue generated by those assets flows back to the vault as yield.
Messari frames InfraFi as one of three viable paths forward for DePIN scaling. The alternatives are capital-light infrastructure with fast payback periods, or leveraging speculative capital during bull markets. For networks requiring significant upfront hardware investment, InfraFi offers a crypto-native debt instrument — effectively securitizing infrastructure revenue streams against stablecoin deposits.
The model is early. Credit risk, duration mismatch, and regulatory classification remain unresolved. But if functional, InfraFi could decouple DePIN growth from token price performance — financing expansion through yield rather than token inflation.
DePIN startups raised approximately $1 billion in 2025, primarily at seed and Series A stages, according to Messari. Between January 2024 and July 2025, $744 million flowed into 165+ startups, with 89 additional deals at undisclosed terms.
Notable fund raises include Borderless Capital's $100 million DePIN Fund III (September 2024) and Entrée Capital's $300 million fund (December 2025). Average fully diluted valuations for newer projects hover around $760 million.
The math is demanding. At $72 million in 2025 sector-wide revenue and $1 billion in venture investment, the capital-to-revenue ratio exceeds 13:1. Even at the $150 million monthly run rate achieved in January 2026, many individual projects remain pre-revenue or deeply subsidized. DePINscan tracks 423 active projects and 41.8 million connected devices — but revenue concentration among fewer than 10 networks suggests the majority may never reach commercial viability.
Revenue concentration. Five networks account for the vast majority of sector revenue. A setback at Aethir or Render would materially impact aggregate figures.
AI cyclicality. GPU compute demand is the primary growth driver. The sector's revenue trajectory is functionally a leveraged bet on continued AI infrastructure spending acceleration.
Token model mismatch. Infrastructure businesses typically require patient, yield-oriented capital. Token models incentivize speculative, momentum-driven capital. The 94–99% drawdown in legacy tokens suggests the market has already priced this structural tension.
Regulatory ambiguity. DePIN networks operate physical infrastructure across jurisdictions. Telecom licensing (Helium), data privacy (Grass), and securities classification (InfraFi vaults) remain largely unaddressed by regulators.
Subsidy dependency. Many networks still rely on token emissions to incentivize supply-side participation. As emissions decline and token prices remain depressed, maintaining network capacity becomes a challenge for projects that have not achieved organic demand.
DePIN in May 2026 presents a sector in transition. The revenue story is genuine — $150 million in monthly on-chain revenue from enterprise customers paying for compute, storage, and bandwidth represents a qualitative shift from the subsidy-driven activity of prior cycles. Aethir's $260 million Axe Compute deal, Render's $38 million January, and Helium's carrier partnerships with T-Mobile and AT&T are commercial milestones that would be notable in traditional infrastructure, let alone crypto.
The counterweight is equally real. Tokens are not recovering. Venture returns are uncertain. Revenue is concentrated in a handful of networks, and the growth engine — enterprise AI compute overflow — is cyclically sensitive. The sector's long tail of 650+ projects remains largely pre-revenue or subsidy-dependent.
What DePIN has achieved is a proof of concept: decentralized networks can generate commercial revenue at scale from enterprise customers. What it has not yet demonstrated is whether token-based business models are the right vehicle for infrastructure businesses that require patient capital, predictable yields, and regulatory clarity. The InfraFi experiment may be the sector's most consequential development — not because it has succeeded, but because its existence acknowledges that the token model alone is insufficient.
The data is clear on one point: someone is paying for the service. Whether the economics work for token holders remains an open question.