Price tells you what the market thinks today. Revenue tells you what users are actually paying for. In crypto, those two numbers often diverge — and right now, the divergence is telling a genuinely interesting story.
The Solana blockchain generated $91 million in monthly application revenue in May 2026 versus Hyperliquid’s $53 million and Ethereum’s $52 million. And for the ninth consecutive quarter, Solana topped all blockchains in decentralised application revenue — a streak running since early 2024.
That is the headline. The context behind it is more interesting than the number.
Three Networks. Three Revenue Models. Three Different Stories.
Solana’s revenue comes from hundreds of applications running on its blockchain — such as swap and lending protocols like Jupiter, Raydium, and dozens of others. Ethereum’s revenue distributes across thousands of protocols. Hyperliquid’s revenue comes almost entirely from one thing: perpetual futures trading on its native DEX.
That concentration is both Hyperliquid’s greatest strength and its most honest structural risk. Hyperliquid leadsall major blockchains with approximately 43% of total weekly blockchain protocol fee revenue in recent weeks — meaning of every dollar paid in fees across all major blockchains combined, roughly 43 cents went to Hyperliquid. This has generated around $11 million weekly. Its fees are driven primarily by perpetuals trading activity, where users pay to open, maintain and close leveraged positions.
Ethereum captures around 13% of fee revenue, derived from a broader mix of DeFi interactions, smart contract executions and token transfers. Solana registersapproximately 10% — a notable gap versus its transaction volume share and a reminder that high-frequency, low-fee trading does not translate efficiently into fee revenue.
The Buyback Flywheel — Why Hyperliquid’s Revenue Structure Is Different
Fee revenue matters differently depending on where it goes. On Ethereum, fees are partially burned — reducing ETH supply. On Solana, fees flow to validators and the Solana Foundation. On Hyperliquid, the protocol funnels 97% of its revenue into an Assistance Fund for HYPE token buybacks (similar to share buy backs by companies). Since launch in 2022, the Assistance Fund has spent over $1.3 billion buying back and burning HYPE tokens — removing approximately 37 million tokens permanently from circulation following a December 2025 validator vote.
Hyperliquid’s cumulative protocol revenue has surpassed$1.027 billion, with an annualised run rate nearing $840 million. Nearly all fees are allocated to ecosystem support and token buybacks, creating consistent buying pressure.
This is a genuinely different economic model from either Solana or Ethereum. It is closer to a listed company buying back its own shares — revenue flows directly to token holders rather than to validators, foundations or burning mechanisms that benefit holders only indirectly.
For context, Hyperliquid’s annualised revenue of approximately $840 million puts it in the same bracket as Bumble, Duolingo and Crocs — globally recognised consumer brands built over many years with thousands of employees. Hyperliquid achieved this with 11 people, no physical offices, and no venture capital. The difference is that virtually all of Hyperliquid’s revenue flows directly back to token holders through buybacks — a distribution model no traditional company offers its customers.
The Honest Counterpoint
Fee comparisons can be misleading, as they reflect different types of on-chain activity. Ethereum’s fees are spread across a vast ecosystem of DeFi protocols, NFTs and decentralised applications, while Hyperliquid’s are concentrated in a single product category.
Solana’s non-memecoin revenue base — trading infrastructure, tokenized equities, stablecoin payments — is the metric to watch heading into Q3 2026 to determine whether its leadership is structurally durable or partly driven by speculative activity.
And Hyperliquid’s concentration in perpetual futures is a real risk. A sustained drop in derivatives demand, increased competition from Coinbase or Robinhood building competing infrastructure, or regulatory action against on-chain perpetuals could compress fee revenue faster than a diversified network would experience.
What the Numbers Actually Tell You
Revenue is the most honest metric available in crypto because it cannot be manufactured. Total Value Locked can be inflated through recursive deposits. Daily active addresses can be gamed. Revenue cannot — it reflects users paying fees for something they chose to use.
By that measure, the current picture is: Solana leads in total application revenue, built on a genuinely diversified ecosystem. Hyperliquid leads in protocol revenue as a percentage of market cap and leads in fee revenue returned to token holders. Ethereum’s fee compression is real and visible in the data — its share of fee revenue has declined significantly even as its total value locked remains dominant.
Three networks. Three different revenue models. All three are generating real economic value. The question for investors is which model they think scales best over the next three years — and that depends entirely on which bet they think is right: broad ecosystem diversification, application-specific specialisation, or institutional settlement infrastructure.
Fee data does not answer that question. But it gives you the most honest starting point available. .
Crypto Finance Asia is Access Asia Group’s crypto and blockchain focused LinkedIn newsletter. Subscribe here.