DePIN Revenue Reality: Which Networks Earn Real Fees in 2026

The Decentralized Physical Infrastructure Network (DePIN) sector's market capitalization grew sharply through 2025, pulling in capital from both retail traders and funds. A project generating real revenue and one surviving on token inflation can look identical on a price chart, which is why the comparison below uses fee data rather than price action.

This distinction matters. Fee revenue may indicate resilience when sentiment turns, since it does not depend on new token buyers the way emissions-funded rewards do. Projects that depend on emissions to reward suppliers need new buyers for those emissions, and when demand for the token slows, the market value of that funding source deteriorates with it, even though the emissions themselves continue on schedule.

Let us examine DePIN revenue project by project. All five figures below are drawn from DefiLlama or the project's own disclosures, but each measures a different thing, detailed under its own heading; none of the five is directly comparable to another. Aethir's headline figure, discussed separately, is not ranked against the other four.

The DePIN Revenue Problem

Most DePIN projects run a simple model: issue tokens as rewards to infrastructure providers, and hope usage catches up before emissions overwhelm demand.

That model creates a fundamental problem. Token emissions paid to node operators, miners, or hotspot hosts are a cost of running the network, not revenue for it. When a project pays suppliers in newly minted tokens, those tokens eventually need buyers. Without real usage generating fees, the arithmetic does not close on any evidence this article has.

We aim for revenue narrowly defined, the way DefiLlama's fee tracker frames it for these networks: money a customer actually pays for storage, compute, or connectivity. That excludes gross "network value transacted," excludes emissions paid to suppliers, and excludes the face value of signed multi-year contracts that have not yet been billed. A protocol can look enormous on the second measure and tiny on the first, and the gap between them is the entire story of this sector. Even among the four DefiLlama-tracked figures below, what each one actually captures differs, as detailed under each project's heading. Aethir's self-reported ARR, discussed below, is furthest from this narrow definition; it is reported alongside the other four projects for reference but is not comparable to them and is never ranked against them.

Helium: A Revenue Story That Changed Mid-Article

Helium (HNT) is the DePIN project most often cited as proof that the sector can generate real revenue. Helium Mobile sells wireless service, and Nova Labs, the company behind it, used a portion of subscriber revenue to buy HNT on the open market through Jupiter DCA and burn it, a mechanism the network's own tokenomics dashboard tracked as a genuine, on-chain buy-and-burn tied to usage.

That total lifetime buyback figure, roughly $2.3 million burned since the program's start, comes from the same DefiLlama tracker. It is a real number, but it is smaller than the figure this article previously cited, and it stopped growing in January. DefiLlama's methodology note also flags a second measurement problem: the Data Credits Helium burns for ordinary network usage are priced at a fixed $0.50-per-gigabyte protocol peg (HIP-149) while carriers actually pay closer to $0.10 per gigabyte, and the HNT used to mint those credits comes from centralized-exchange treasury wallets rather than an open-market purchase. Neither of those two burn types is a clean proxy for what customers are paying.

None of this means Helium Mobile has no subscribers or no revenue. It means the on-chain, verifiable signal that made Helium the DePIN sector's favorite revenue story went dark in January 2026, and any figure claiming otherwise now needs a source newer than that date.

Render Network: Real Fees, Modest Scale

Render Network (RENDER) sells GPU compute: users pay in USD terms for rendering and AI inference jobs, and node operators supply the hardware. Its Burn-Mint Equilibrium model (RNP-001) burns RENDER equal to 95% of each job's price on the Solana chain, with the remaining 5% paid off-chain to OTOY as a service fee.

Measured through DefiLlama's fee tracker for Render, the network generated roughly $156,000 in gross job fees over the trailing 30 days, an annualized run rate near $1.9 million. That is real, verifiable customer spend, tracked from on-chain burns rather than a self-reported figure.

Render has expanded past 3D rendering into general AI compute through its RNP-019 subnet, and partnerships with firms including Stability AI point to genuine commercial interest beyond crypto-native rendering demand. The obstacle is competitive pricing: AWS, Google Cloud, and Azure already sell GPU time with service level agreements and existing enterprise relationships Render has not yet matched at scale.

Filecoin: Capacity Without a Matching Fee Line

Filecoin (FIL) is the largest decentralized storage network by capacity. Checked against Filfox's live network explorer, the network's total adjusted storage power stood at roughly 12.4 exbibytes as of this article's 2026-08-16 update; a year-earlier report put it at 3.0 exbibytes, though whether both figures measure the same metric is unverified, spread across more than 500 active mining operators.

Filecoin's transaction-fee revenue has stayed modest while its reported storage figure grew from 3.0 to 12.4 exbibytes; whether both endpoints measure the same metric, raw versus quality-adjusted power, is not something this article verifies. DefiLlama's fee tracker for Filecoin puts the network's total transaction-fee revenue at roughly $221,000 over the trailing 30 days, an annualized run rate near $2.7 million, against a lifetime total of $5.3 million recorded since tracking began. Filecoin's own token has traded far below its all-time high through this period.

Filecoin's situation shows how hard this sector's core question is to answer from public data: build the infrastructure first, and hope commercial demand for it follows. On the explorer's numbers the reported storage figure is roughly 4x higher year over year, a comparison this article cannot fully validate, since it cannot verify that both endpoints measure the same metric; no comparable year-earlier fee figure is available here to check whether transaction-fee revenue grew at anything close to that rate, and that figure says nothing about storage-deal revenue specifically, which this article does not have a number for.

Aethir: Enterprise Compute, With a Contract-Value Caveat

Aethir self-reported $141 million in annualized recurring revenue (ARR) as of mid-2025 in an H1 2025 report that is no longer reachable at its original URL, so the figure is unverifiable here, and it was never comparable to or ranked against the other four projects in this piece. The report also stated 955 million compute hours delivered across more than 430,000 GPU containers in 94 countries, and over 150 new partners and clients onboarded in the first half of the year.

Unlike projects built around retail node operators, Aethir sells to enterprise customers directly, producing larger contracts with a different risk shape than a consumer subscription model.

That said, ARR is a run rate calculated from current contracts, not cash already collected, and Aethir's own newer disclosures make the distinction sharper. Its enterprise arm, Axe Compute, announced in mid-2026 that it had signed more than $3 billion in total contract value for the year, including one five-year, $1.5 billion deal for over 9,200 NVIDIA Blackwell B300 GPUs. The same announcement puts expected customer prepayments at roughly $534 million and the projected annual run rate once those contracts are fully deployed at above $696 million, explicitly still ahead of actual deployment and billing. Signed contract value, prepayments, and recognized revenue are three different numbers, and Aethir's own reporting keeps them separate. Readers comparing DePIN projects should keep them separate too.

The trade-off behind the enterprise model is concentration risk: a handful of large customers can move the revenue line sharply in either direction, in a way thousands of small retail subscribers cannot.

Comparing DePIN Revenue Models

The four DePIN revenue figures below are all independently tracked by DefiLlama, but each one measures a different thing, so this is a per-project summary rather than a like-for-like ranking. Helium's number is a paused buyback burn, not current subscriber payments. Akash's and Render's numbers are fees paid directly for the compute service. Filecoin's number is transaction gas, not storage-deal spend. Aethir is listed separately underneath because its headline figure is not built the same way and is not ranked against these four.

Aethir (not comparable to the table above): Enterprise compute. $141M self-reported ARR, a forward-looking run rate calculated from current contracts, not recognized revenue or cash already collected (H1 2025, Aethir's own report). No independent tracker publishes an Aethir fee figure to check it against, so it is reported here for reference only and is never ranked against Helium, Akash, Filecoin, or Render.

Among these DePIN revenue figures, Helium's on-chain signal is currently the least useful, not because Helium Mobile has no subscribers, but because the mechanism that made its revenue verifiable stopped running in January 2026. Akash's and Render's numbers are the closest of the four to a direct fee-for-service measure. Filecoin's figure describes chain usage, not the storage-deal spend the sector's marketing usually implies.

The Token Burn Question

Revenue generation is only half the equation. What a project does with that revenue determines whether token holders benefit from it.

Akash Network burns AKT proportional to compute spend under its Burn-Mint Equilibrium (BME) model, the same underlying mechanism Render uses. DefiLlama measures Akash's fee revenue at roughly $231,000 over the trailing 30 days, a figure the tracker attributes to "lease fees paid by users to use Akash Network services," with burned coins recorded as the network's revenue line.

Render burns RENDER equal to 95% of each job's price on-chain and routes the remaining 5% off-chain to OTOY, so most of each job's dollar value is removed from circulation. Whether that translates into net deflation depends on the mint side of the Burn-Mint Equilibrium, the emissions paid to node operators, which this article does not quantify.

Projects still emitting new supply without a clear, working burn or accrual mechanism face ongoing dilution regardless of revenue growth. Emissions can outpace fee generation even at a network that is nominally growing its business, and Helium's paused buyback shows that a burn mechanism can also simply stop, which is a distinct risk from never having built one.

DePIN Revenue: Investment Implications

For anyone evaluating DePIN projects, revenue metrics have to be evaluated alongside network usage, emissions, and longitudinal fee growth, but they are the input the sector's marketing most often skips.

Strong signals:

• A revenue figure sourced to an independent tracker or an audited disclosure, not a marketing deck
• Fee growth that keeps pace with, or outpaces, token emissions
• A token burn or revenue-share mechanism that is currently active and verifiable on-chain
• Enterprise contracts disclosed with a clear distinction between signed value and recognized revenue
• Storage, compute, or bandwidth capacity that is actually being paid for, not merely built

Warning signs:

• Revenue claims that cite "network value transacted" or gross activity instead of fees paid
• A burn mechanism the project has quietly paused or reduced without an update to its public materials
• Contract value or ARR presented without the underlying deployment or billing status
• Capacity or node-count growth with no corresponding fee-revenue growth over the same period
• No independently verifiable, per-protocol revenue figure at all

The DePIN sector contains real infrastructure and real engineering. The current and 30-day annualized fee figures gathered here are one input for judging financial sustainability, not longitudinal survival evidence; they may point toward a defensible fee base, but they do not by themselves show which networks needed a bull market to survive.

For more on how AI agents are driving infrastructure demand, or how Solana's ecosystem supports DePIN projects, see our related coverage.

The DePIN Revenue Bottom Line

DePIN revenue reality, checked against DefiLlama and each project's own disclosures, is more mixed than the sector's marketing suggests. Akash, Filecoin, and Render each show verifiable, independently tracked fee revenue in the low millions annualized. Helium's revenue story, the sector's most-cited example, lost its on-chain verification in January 2026 when the buyback program that made it measurable was paused. Aethir reports a separate, not-comparable figure, $141 million self-reported ARR, plus more than $3 billion in signed contract value for 2026, most of which is still ahead of deployment and billing; neither number is ranked against the other three projects' independently tracked fee revenue.

Capacity, node counts, and transaction volume cannot establish financial sustainability without a fee line customers are actually paying into. Investors evaluating DePIN tokens should ask for that fee line, sourced and dated, before anything else.

Disclaimer: Nothing in this piece is financial advice. Cryptocurrency investments carry significant risk. Always conduct independent research and consult a qualified financial advisor before making investment decisions.