Decentralized physical infrastructure networks, commonly called DePIN, are the category of crypto projects that coordinate real-world hardware through token incentives. A device owner installs a sensor, router, camera, or storage drive. The network pays them in tokens for contributing data or capacity. Enough devices join and you have infrastructure that no single company controls.
That pitch has been around since Helium launched in 2019. What changed in 2023 and 2024 was the number of projects shipping working products rather than roadmaps. By 2025, DePIN had become one of the more substantive verticals in crypto — not by price performance, but by actual deployed hardware and measurable network activity.
The Protocols That Matter
Helium is the origin story. The network started as a decentralized wireless provider for IoT devices, and at its peak operated hundreds of thousands of Hotspot nodes. When IoT demand failed to materialize at the scale the tokenomics needed, Helium pivoted. It migrated to Solana in April 2023 and relaunched with a mobile carrier model, partnering with T-Mobile to offer discounted plans routed through Helium Mobile. Subscriber numbers remain small, but the model is coherent: token rewards subsidize coverage, and coverage sells plans.
Hivemapper coordinates a dashcam network. Drivers install a Hivemapper dashcam, collect street-level imagery, and earn HONEY tokens for contributing to a global map. The network has mapped tens of millions of kilometers of roads, with particularly dense coverage in markets underserved by Google Street View. The data has commercial buyers. That is more than most DePIN projects can say.
io.net aggregates underutilized GPU capacity from data centers, crypto miners, and consumer hardware into a marketplace for machine learning compute. It launched on Solana in 2024 and positioned itself as a cheaper alternative to centralized cloud GPU providers for AI inference and model training workloads. Demand from AI developers is real; whether the supply side remains coherent as GPU market conditions shift is the open question.
DIMO connects vehicles to a decentralized network. Car owners install a hardware device or use a compatible car’s data connection. The network collects telematics data — mileage, location, diagnostics — and the owner controls what gets shared with insurers, fleet operators, or researchers. DIMO’s pitch is data portability: your car generates valuable information, and under the current model, automakers and dealers capture it. DIMO proposes to route that value back to the owner.
The Token Problem
Every DePIN project faces the same structural tension. Hardware deployment is expensive and slow. Token inflation to incentivize deployment is cheap and fast. In the early stage, token rewards subsidize supply-side growth. The risk is that supply growth outpaces demand, which pushes token prices down, which reduces the real-value reward for operators, which causes supply to contract.
Helium ran into this directly. IoT data revenue never covered the cost of Hotspot subsidies. The network survived because it pivoted its use case, not because the original model worked.
The DePIN projects with the most durable economics are the ones where token rewards are backstopped by real revenue: data buyers, service subscribers, compute customers paying in USD. Hivemapper has commercial map data purchasers. io.net has AI developers paying for GPU time. Those revenue streams reduce but do not eliminate the dependency on token appreciation.
What the Data Shows
Messari’s annual DePIN report, published in early 2025, estimated the combined service revenue across major DePIN networks was still a fraction of the token inflation used to incentivize supply. That gap has to close for the category to be self-sustaining. In 2025, several projects began moving toward revenue-share models, where operators receive a portion of actual network revenue rather than purely inflationary token emissions.
The on-chain activity tells a partial story. Networks like io.net and DIMO show consistent transaction volumes, which reflect real device interactions rather than speculative trading. Helium Mobile’s subscriber base grew throughout 2025. Filecoin, the oldest decentralized storage network, saw storage utilization tick up as enterprises tested it for archival use cases.
None of these numbers suggest DePIN has cracked the infrastructure market. They do suggest the category is past pure speculation and into the phase where the real-world performance of each network determines which projects survive.
The Watchlist
Three things to track over the next twelve months. First, whether io.net can retain GPU supply as centralized cloud providers cut prices in response to new chip supply. Second, whether Helium Mobile reaches a subscriber threshold that makes the mobile carrier model financially viable without heavy token subsidy. Third, whether any DePIN project closes a material enterprise contract that is publicly verifiable — that would be the category’s proof of institutional legitimacy.
DePIN is building real things. The infrastructure exists and in some cases generates real revenue. The question is whether the token model can transition from growth subsidy to sustainable incentive before the inflation costs price operators out.