
Gas Is Becoming Obsolete: From VM to Resource Market, Blockchain Is Moving Toward the "Chain Cloud"
By Fu Gui
The notion that "Gas is becoming obsolete" does not imply that Gas will vanish anytime soon, but rather that its explanatory power as a unified resource abstraction for blockchains is diminishing. This judgment does not stem from a technological upgrade on a single chain, but rather from four entirely different directions, all converging on the same outcome.
Four Directions, One Signal
The first direction comes from Hyperliquid. On this decentralized perpetual contract exchange, core users routinely see maker and taker trading fees, rather than Gas multiplied by Gas Price. The HyperEVM still uses HYPE as Gas, and EIP-1559's base fee and priority fee are also operational, but the transaction layer has wrapped underlying resource costs into trading service fees. When a user opens a BTC long position, what they are purchasing is matching and settlement, not computational instructions within block space. Gas here is not eliminated; it is merely hidden within business pricing.
The second direction comes from Solana. SIMD-0553 once entered accepted and merged states, but the accompanying governance proposal SGP-0003 was rejected in August 2026, and the resource fee model has not yet been deployed on mainnet. Nevertheless, the split it proposes—the Base Inclusion Fee and the Resource Fee based on requested cost units—clearly indicates that Solana is decoupling the pricing of "being included" from "consuming resources." The question the Base Inclusion Fee answers is "Is this transaction worth including in a block?", while the Resource Fee answers "Exactly how many network resources does this transaction consume?" At the foundation of the Resource Fee are requested cost units, which explicitly list signature cost, write lock cost, instruction data cost, program execution cost, and loaded account data cost. Solana itself is breaking Gas back down into those concrete resources it originally concealed.
The third direction comes from ICP. The Internet Computer has always used cycles instead of Gas; cycles are pegged to the XDR and measure real resources such as compute, storage, bandwidth, and messaging. Launched in 2026, Cloud Engines takes this a step further: users can select nodes, node providers, geographic locations, and replication scales, allowing the protocol to assemble a dedicated execution environment. Officially referred to as a user-owned slice of Internet Computer capacity. When users are purchasing a compute environment rather than a single transaction, the concept of Gas itself recedes into the background.
The fourth direction is the most notable because it comes from Ethereum itself. On August 17, 2026, Ethereum launched the Platåberget testnet as an early public testing environment for the Glamsterdam upgrade. Officials warned in advance: wallets, indexers, and gas estimators that rely on fixed Gas Limit assumptions may all be disrupted by the upgrade. Glamsterdam adjusts the Gas cost for operations like state access; EIP-8037 and EIP-8038 respectively increase and partially independently meter the costs of state creation and state access, making Gas more aligned with the actual workload of nodes. Meanwhile, EIP-7999 is advancing a unified multidimensional fee market. Ethereum itself is also breaking Gas back down into what it originally concealed: state access, computation, data, and various types of node work.
Viewed together, these four trajectories lead to a clear conclusion. Gas is retreating from being a core blockchain abstraction to becoming the underlying settlement mechanism of the resource market.
The Problem May Not Be Gas at All, But the VM
Since Gas is being dismantled, the industry's focus on VMs over the past decade may have been misguided from the start.
The EVM remains the ecosystem's most powerful smart contract execution environment, and the network effect built around Solidity, OpenZeppelin, Foundry, MetaMask, Uniswap, and Aave still stands unrivaled. Yet as the "underlying execution model of the World Computer," the EVM is increasingly struggling to bear the full burden of the next generation of global verifiable computing infrastructure on its own. The issue is not that it cannot run programs, but that it conflates execution, state, resources, and proofs into a single abstraction.
From an execution model perspective, the EVM is a Stack Machine where each stack item is 256-bit. Modern CPUs prefer a register architecture, whereas every ADD and MUL operation in the EVM is accompanied by extensive stack manipulation. FuelVM's adoption of 64-bit registers is itself a direct response to this historical baggage.
From a parallelization model perspective, EVM transactions do not declare state dependencies, leaving the execution layer unable to schedule them in advance. Whether two transactions access the same state is nearly invisible within the EVM. Solana's Sealevel transforms state dependencies into something the scheduler can anticipate upfront: transactions declare which accounts they need to access, the Scheduler determines dependency relationships, and then executes them in parallel. This philosophy closely mirrors dependency graphs in modern CPUs. Solana's true revolutionary aspect lies not in the sBPF instruction set, but in elevating "who can run in parallel" from a hidden runtime state to a first-class citizen in the execution model.
From a proving model perspective, the EVM's 256-bit stack structure and state access pattern are inherently unfavorable for ZK proving. zkEVMs have long suffered immensely from proving costs, yet future chains must assume that execution will be followed by proof generation. RISC-V grows ever more important precisely because it grants blockchains their first near-CPU-level underlying execution standard, allowing ecosystems like LLVM and GCC, register architectures, and zkVM standardizations to seamlessly integrate. Ethereum and Polkadot both moving toward RISC-V simultaneously is no coincidence.
Move's contribution exists on an entirely different dimension. Rather than chasing "faster opcodes," it transforms assets from data within smart contracts into language-level resources. In Solidity, double spending relies on audits and libraries for prevention, whereas Move directly enforces this constraint through its language and type system. Concepts like resource safety, ownership, and abilities address not execution speed, but resource security management.
Therefore, the VM competition itself is a false dichotomy. EVM offers composable generic smart contracts, SVM provides explicit dependencies and parallel scheduling, Move delivers resource safety, RISC-V supplies a universal execution ISA, and zkVM enables provable execution. In the future, no single VM will win outright; what truly gets reinvented is the entire computer.
Blockchain Took Ten Years to Rediscover "Resources"
Blockchain did not invent resource pricing for the first time. It simply took ten years to rediscover what the computer industry knew decades ago: computation is not a single resource, but a suite of resources.
Ethereum's classic model relies on single-dimensional Gas, where Fees equal GasUsed multiplied by GasPrice. While extremely simple, this model harbors a fundamental economic flaw: different resources are non-fungible, yet priced under a single metric. A transaction with high CPU usage but low storage requirements might be tagged at 100 Gas, identical to one with minimal CPU but massive storage needs. This price tag fails to accurately reflect a node's true bottleneck.
EOS began implementing three independent resources back in 2018: CPU, NET, and RAM. RAM represents property rights that can be purchased, used, and resold, with prices dynamically shifting via the Bancor algorithm. CPU and NET are leased resources, later evolving into the PowerUp model, which essentially rents compute and network resources for a set duration. EOS differentiated between Ephemeral Resources and Persistent Resources early on, predating Sui's separation of Computation and Storage by several years.
TRON follows a different path. Bandwidth corresponds to transaction bytes, Energy to contract execution, and TRON Power to governance. Users obtain resource quotas by staking TRX and can delegate these resources to others. TRON even allows DApp developers to configure the consume user resource percent parameter, directly absorbing Energy costs on behalf of users. This aligns closely with later developments like Paymasters and sponsored transactions.
Hive completely abandoned Gas, adopting a conversion from HP to RC. Resource Credits are not tokens; they are renewable credit quotas granted to accounts based on Hive Power. RC regenerates automatically after consumption over a five-day cycle. Essentially, this operates on a quota model, well-suited for social and content applications but ill-fitted for complex DeFi and high-frequency computation.
These legacy chains solved "resource metering" but fell short on "resource markets" and "resource abstraction." Users still needed to comprehend concepts like CPU, NET, RAM, Energy, and Bandwidth to interact with the chains normally. This lack of intuitive UX is one reason they ultimately failed to become the standard answer. Yet their paradigms never disappeared. Solana isolated computation into CUs, turning transactions into resource requests. Sui split fees into Computation and Storage, being the first mainstream L1 to explicitly distinguish "executing once" from "permanently storing 1MB" as entirely separate economic actions. Polkadot constructs a two-dimensional Weight using ref_time and proof_size, supplemented by storage deposits to manage long-term state growth. The shift from Gas to multidimensional resources isn't about new chains being smarter than old ones; it reflects the industry's gradual acknowledgment that blockchain nodes consume far more than one type of genuine resource.
The Real Battle Is Resource Pricing
Resource metering is merely the first step. The core challenges that remain are: who sets the prices for these resources? How are they allocated? Can they form a functional market?
Hedera offers a compelling answer. It decomposes transaction costs into Node fee, Network fee, and Service fee, weighted behind the scenes by Bandwidth, Processing, Storage, and Duration, assigning higher prices to scarcer resources. This moves beyond Opcode Gas Accounting into true Resource Cost Accounting. More intriguingly, Hedera rolled out Simple Fees in 2026, packaging complex resource rates into a Base plus Extras structure. Multidimensional under the hood, single-dimensional at the user interface—this is highly viable engineering-wise.
ICP pushes this further. By pegging cycles to the XDR, explosive token price surges do not force server resource costs to spike in tandem; developers face relatively stable fiat-denominated resource prices. Compute Allocation bills directly based on core utilization ratios, pricing one percent of a core at a fixed rate per second in cycles. This isn't a transaction fee; it's a cloud computing bill.
Filecoin demonstrates the purest form of a resource market. Clients negotiate prices directly with storage providers, broadcasting transactions on-chain to form a Storage Market. Prices are not protocol-mandated but determined directly by supply and demand. This stands as the closest case in the industry to a decentralized resource market.
Beyond this, block space includes another often-overlooked resource dimension: temporal priority (Inclusion Priority and Ordering Rights). Priority fees, Jito-style auctions, and Execution Tickets are essentially placing value on "who gets included first and in what order." If Compute, Storage, and DA correspond to a node's physical and verification costs, then Priority maps to scarce positioning within blocks. MEV can be viewed as a crude, fragmented valuation of this special asset; a mature resource market should explicitly integrate it rather than let it linger outside the protocol indefinitely.
Drawing from these implementations, we can abstract four tiers: Resource Accounting solves "how much was consumed", Resource Pricing solves "what is the cost", Resource Allocation solves "who gets the resources", and Resource Market solves "how resources trade freely". Future blockchain pricing will not be about "better Gas," but rather a multidimensional resource pricing framework. Fees will equal the sum of all resource consumption multiplied by their respective prices, encompassing Compute, State, Storage, Bandwidth, DA, Proof, Concurrency, Messaging, and Priority.
ChainCloud: When Blockchain Begins to Revert to Cloud
The World Computer has never quite felt like a Computer, not due to a lack of fast enough VMs, but because it possesses only "execution" without complete resource management, scheduling, pricing, isolation, and markets. A genuine computer would not let each program define its own CPU price. A true cloud would never ask users to calculate how much RAM they are buying right now.
ICP Cloud Engines marks the first true productization of "blockchain plus cloud". Users select nodes, providers, geographic locations, and replication scales, while the protocol assembles a dedicated execution environment. The official Dashboard already displays data centers, node providers, and machine specifications. With measured restraint, this does not automatically prove ICP is the final answer. What it offers is a prototype of ChainCloud worthy of observation: the ability to organize nodes, hardware, location, replication, and execution environments into a cohesive cloud service.
Hyperliquid and GameFi validate another trajectory. Hyperliquid users purchase trading services, while GameFi players purchase gaming experiences. Immutable's Orderbook enables fully Gasless NFT listings upon user signing, separating the Gas payer from the transaction signer. Gas shifts from a user expense to a platform operating cost. This perfectly mirrors the Web2 economic model: users pay for products, not infrastructure.
ChainCloud is not AWS plus Tokens. It is a globally verifiable compute resource pool managed by protocols, priced by markets, provisioned by nodes, secured by cryptography, and connected by unified execution standards.
The Next Generation Blockchain Is a Single Machine

Pushing this logic to its conclusion, the structure of the next-generation Blockchain is no longer a "single chain," but rather a stacked five-layer architecture.
At the top sits the Service Layer. Applications like Trading, Gaming, Storage, and AI face users directly, presenting business prices rather than resource prices.
Below that lies Resource Abstraction. This layer packages underlying multidimensional resources into service interfaces comprehensible to upper layers. Ethereum's EIP-7999 has already architected this direction: multidimensional underneath, single-dimensional at the UI level; users simply provide a unified max fee, while the protocol internally allocates the budget across different resources.
Further down is the Resource Market. Compute, State, Storage, DA, Bandwidth, Proof, and Priority each carry independent prices dictated by supply and demand, forming tradable resource pools.
Beneath the Resource Market sits Parallel Runtime. This layer handles dependency graph analysis, concurrent scheduling, and state access management. Sealevel, Block-STM, and the Sui Object Model all belong to this tier.
At the very bottom are Execution ISA and Distributed State. RISC-V resides here, serving merely as a universal execution standard beneath ChainCloud, not the core itself. The true core lies in the combination of ISA, Runtime, State, Resource Market, and Proof.
End Users Should Never Have to Know About Gas
In the future, users should never encounter terms like Gas, Gwei, CU, ref_time, proof_size, or storage deposit. They should only see "Swap 100 USDC for ETH", "Buy a sword", "Store 1GB for 100 years", or "Open a BTC long position".
The underlying pricing formula is multidimensional: Fees equal the consumption of each resource multiplied by its current price. But the user-layer pricing formula diverges completely: User Price equals Service Value plus Resource Cost, plus Risk Premium, plus Platform Margin. The foundation meters resources with increasing precision, while the ceiling conceals them with growing thoroughness. This is precisely how a truly mature computer should operate.
After the World Computer
Gas is not dead; it is stepping down. It has transformed from a raw resource into a packaging layer for resource markets.
Blockchain's evolutionary path is gradually coming into focus: from Blockchain to World Computer, to Resource Computer, to Resource Market, to Decentralized Cloud, culminating in Chain-Cloud. Chains solve trust; clouds solve resource allocation. The mission for next-generation blockchains is not to migrate all computation on-chain, but to transform global compute resources into verifiable, tradable, programmable public infrastructure. Once this transition completes, blockchain will cease to be merely a faster chain and will instead become a true, cryptographically secured cloud.














