Rollup-as-a-Service (RaaS) Guide 2026
RaaS platforms have transformed how teams launch custom blockchain rollups. Instead of spending months building infrastructure, you can now deploy a production-ready chain in days. We compare Conduit, Caldera, Gelato, and AltLayer—the leaders shaping the 2026 RaaS landscape.
1. What Is Rollup-as-a-Service?
Rollup-as-a-Service (RaaS) is a managed infrastructure platform that enables teams to deploy custom Layer 2 or Layer 3 rollup chains without building infrastructure from scratch. Think of it as AWS for blockchains: you describe what you want (throughput, data availability layer, execution framework), and the provider handles the operational complexity—sequencers, RPC nodes, bridge infrastructure, state storage, and upgrades.
The RaaS market has grown explosively. In 2024, the RaaS market was valued at approximately $75 million; projections show it reaching $354 million by 2032—a 4.7x CAGR. This reflects the industry's shift from "teams build their own chains" to "teams deploy managed chains in minutes." With over 100 rollup chains now live on Ethereum alone, the infrastructure commoditization is complete.
Before RaaS, launching a rollup required: deep expertise in Layer 2 design, a DevOps team to manage sequencers and full nodes, time to integrate bridges and liquidity, and constant vigilance against infrastructure failures. Now, RaaS abstracts these away. A five-person project team can deploy a fully operational chain with custom throughput, security guarantees, and tokenomics.
💡 The RaaS Value Proposition
Speed: Deploy a mainnet-ready rollup in days to weeks, not months. Cost: Avoid hiring infrastructure engineers; pay only for what you use. Uptime: Leverage provider's 99.9%+ SLAs and DDoS protection. Modularity: Choose your data availability layer (Ethereum, Celestia, EigenDA) and execution framework independently. Focus: Spend time on product and tokenomics, not sequencer management.
RaaS providers typically offer a modular stack—you pick your execution layer (OP Stack, Arbitrum Orbit, ZK Stack), data availability (Ethereum blobs, Celestia, EigenDA), and optional sequencing model. Most chains launched via RaaS in 2024-2026 use Ethereum as settlement, Ethereum blobs or Celestia for DA, and a provider-managed sequencer for simplicity and security.
2. Why RaaS Matters in 2026
The modular blockchain thesis has matured. When rollups first emerged (2021-2022), most teams deployed on two monolithic chains: Ethereum (for security) and a custom sequence layer. Today, the stack is fully decomposed—execution, settlement, data availability, and sequencing are all separable concerns. RaaS providers emerged to abstract away this complexity.
The proliferation of rollups is the clearest sign RaaS is working. As of March 2026, there are 100+ live rollup chains—Optimism, Arbitrum, Degen, Zora, Base, Blast, Mantle, Linea, and dozens more. A year ago, deploying a new chain was still a 6-month engineering effort. Now, serious projects view custom rollups as standard infrastructure, not experimental luxury. Uniswap, Aave, and major DAOs are exploring or running their own rollups.
RaaS unlocks several critical capabilities:
Custom Execution Environments
You don't need Ethereum's full VM. Gaming chains can optimize for low-latency gaming loops; finance chains can add custom precompiles for derivatives. RaaS lets you fork or customize the VM without managing your own proving system.
Economic Customization
Set your own gas fee structure, MEV auction mechanisms, and token allocation. Unlike using Ethereum or Arbitrum, your rollup's economics are entirely yours to design.
DA Cost Reduction
Ethereum blobs reduced DA costs 100x vs. calldata. Celestia and EigenDA go further, with custom DA layers optimized for rollups. RaaS lets you pick the cheapest DA that fits your security model.
Bridging Infrastructure
RaaS providers handle canonical bridge infrastructure, liquidity provisioning partnerships, and bridge monitoring. You don't maintain bridge code.
The "appchain" movement—where applications deploy their own rollups for customization and brand control—is now possible because of RaaS. Without it, only the wealthiest DAOs and protocols could afford custom chains. With RaaS, any mid-size protocol can explore a rollup roadmap.
3. The RaaS Tech Stack
A complete rollup stack consists of five layers. RaaS providers let you mix and match across each layer, or provide opinionated presets for common use cases:
| Layer | Examples | Purpose |
|---|---|---|
| Execution Framework | OP Stack, Arbitrum Orbit, ZK Stack, Polygon CDK | Defines rollup type (optimistic/zk) and VM (EVM/custom) |
| Settlement Layer | Ethereum mainnet, Celestia, Avail | Where your rollup posts commitments and can be settled |
| Data Availability | Ethereum blobs, Celestia, EigenDA, Avail, NearDA | Stores transaction data for light clients to reconstruct state |
| Sequencing | Provider-run, MEV-auctions, AVS (AltLayer) | Orders transactions and creates blocks |
| Bridging | Native optimistic/zk proofs, Hyperlane, LayerZero | Moves assets between rollup and settlement/other chains |
📊 Typical 2026 Stack Configuration
Execution: OP Stack (optimistic rollup, widely adopted) or Arbitrum Orbit (AnyTrust protocol, faster finality).
Settlement: Ethereum mainnet (most secure, standard).
Data Availability: Ethereum blobs (cheap, native), or Celestia (ultra-cheap, modular).
Sequencing: Centralized (provider-run for initial launch), or decentralized via AVS (AltLayer) for longer-term decentralization.
Bridging: Native optimistic proofs, or third-party aggregators like Hyperlane for interop with other chains.
The critical innovation of RaaS is modularity. Historical rollups (early Optimism, Arbitrum) were tightly coupled—you couldn't change the DA layer without redeploying. Modern RaaS stacks decouple everything, letting you optimize cost and performance independently.
4. Top RaaS Providers Compared
Four providers dominate the RaaS market in 2026: Conduit (simplicity-first), Caldera (flexibility-first), Gelato (infrastructure-complete), and AltLayer (decentralization-first). Here's how they stack up:
| Provider | Frameworks | DA Options | Key Feature | Notable Users |
|---|---|---|---|---|
| Conduit | OP Stack, Arbitrum Orbit | Ethereum blobs, Celestia, Avail, Eigenda | No-code deployment, 3-step launch, highest uptime SLA | Zora, Aevo, Degen, Synthetix (Andromeda) |
| Caldera | OP Stack, Arbitrum Orbit, ZK Stack, Polygon CDK | Ethereum blobs, Celestia, EigenDA, Avail, NearDA | 40+ infrastructure integrations, maximum customization, preset chains | Manta Pacific, Sanko, Evernode |
| Gelato | OP Stack, Arbitrum Orbit, custom | Ethereum blobs, Celestia, EigenDA | Full-stack managed infra, auto-scaling RPC, web3 services (Relay, Functions) | Thirdweb, Web3.Storage ecosystem |
| AltLayer | OP Stack, Arbitrum Orbit with AVS | Ethereum blobs, Celestia, custom DA via AVS | Restaked rollups, decentralized sequencing (MACH/SQUAD/VITAL), Eigenlayer integration | Xterio (gaming), DODOchain (DeFi) |
Conduit: Simplicity First
Conduit is the no-code RaaS platform. You answer a few questions on their dashboard (What throughput do you need? Ethereum or Celestia for DA? Custom token?), and 3 steps later, you have a testnet rollup. Mainnet deployment is similarly frictionless. Conduit focuses on the OP Stack and Arbitrum Orbit, and abstracts away technical decisions to reduce decision paralysis.
Conduit's playbook: sensible defaults, 99.99% uptime SLA, built-in faucets, explorers, and RPC endpoints. They've launched chains for Zora (NFT-focused), Aevo (derivatives), Degen (meme tokens), and Synthetix's Andromeda chain. Pricing is transparent—you pay per blob or a fixed monthly fee depending on DA choice.
Best for: Teams that want a rollup ASAP without customization headaches. Gaming projects, community chains, and brand-driven chains often choose Conduit because speed matters more than VM customization.
Caldera: Customization King
Caldera emphasizes flexibility. They support OP Stack, Arbitrum Orbit, ZK Stack (Cairo-based), and Polygon CDK (zkEVM). More importantly, Caldera integrates 40+ third-party services—custom data availability layers (EigenDA, Avail, NearDA), sequencing solutions, bridge aggregators, and block explorers. You can assemble a highly custom stack tailored to your throughput and cost requirements.
Caldera also offers preset "Caldera Chains"—preconfigured rollups optimized for specific use cases (gaming, DeFi, NFTs). But the real power is the ability to swap components. For example, if you want ZK proofs but custom DA, Caldera makes that possible. They've launched chains for Manta Pacific (privacy-focused), Sanko (gaming), and Evernode (infrastructure).
Best for: Teams with specific technical requirements—custom VMs, exotic DA layers, or unique tokenomics. Enterprise projects and DAOs with engineering teams.
Gelato: Full-Stack Infrastructure
Gelato takes a different angle: managed operations. Beyond rollup deployment, Gelato provides auto-scaling RPC nodes, custom smart contract execution (Gelato Functions), MEV protection tools (Relay), and monitoring/alerting. Think of Gelato as "managed Web3 infrastructure at scale."
Gelato's strength is in reducing operational overhead. You deploy a rollup, and Gelato manages RPC scaling automatically as traffic grows—no manual node management. Their Relay product helps you avoid MEV extraction, and their Functions product lets you run off-chain code triggered by on-chain events. They've built rollup infrastructure for Thirdweb's ecosystem and other Web3 platforms needing white-glove support.
Best for: Teams that value operational peace-of-mind and want a full suite of Web3 infrastructure tools beyond just rollup deployment.
AltLayer: Restaked Rollups & Decentralization
AltLayer brings a unique concept: restaked rollups. Instead of relying on a centralized sequencer (the default for other RaaS), AltLayer enables decentralized sequencing via Eigenlayer's AVS (Actively Validated Services). Multiple operators validate your sequencer's correctness and are slashed if they behave maliciously. This gives you decentralization closer to Ethereum's validator set, without running your own consensus.
AltLayer's tech stack includes MACH (MEV-Aware Consensus with Helix), SQUAD (Sequencer Quorum Agreement Delegated), and VITAL (Validator-Infused Transaction Aggregation Layer)—different sequencing models with varying decentralization/latency tradeoffs. Tokens staked in these AVS get yield from your rollup's MEV and sequencing fees.
AltLayer is live chains: Xterio (a gaming-focused rollup with EVM compatibility), DODOchain (DeFi), and others. The model aligns incentives—your sequencer's security is backed by restaked validators who earn yield.
Best for: Projects that prioritize decentralized sequencing and want to bootstrap a validators community around their chain. Gaming and DeFi projects where MEV and sequencing integrity matter.
5. How to Choose a RaaS Provider
Choosing a RaaS provider is a major decision—you'll be locked in for the life of your chain. Here's a framework to evaluate each provider against your project's needs:
Throughput Requirements
Do you need 1,000 TPS or 10 TPS? Conduit and Gelato scale easily; Caldera gives you tools to optimize for your specific throughput. Lower throughput = lower DA costs, so don't over-provision.
Data Availability Cost Sensitivity
Ethereum blobs are cheapest (post-Dencun); Celestia is 10x cheaper again. If DA is 50%+ of your costs, swap DA layers. Caldera's flexibility here is unmatched.
Framework Preference
OP Stack is most battle-tested; Arbitrum Orbit has faster finality; ZK is emerging. Conduit is OP-centric; Caldera supports all. Know what you need before you choose.
Decentralization Timeline
If you want decentralized sequencing day-one, pick AltLayer. If you're comfortable with centralized sequencing initially, others offer paths to decentralization later.
Time-to-Launch
Conduit: 2 weeks. Caldera: 4 weeks with customization. Gelato/AltLayer: 3-6 weeks with full integration. If you need launch in 10 days, Conduit is your only option.
Ecosystem & Liquidity
Chains deployed via Conduit (Zora, Degen, Base) have active DEX ecosystems. Newer providers are still building liquidity. If DEX availability on day-one matters, established providers win.
🎯 Decision Tree
Fastest launch + sensible defaults? → Conduit
Maximum customization + exotic tech? → Caldera
Full-stack Web3 infra + ops peace? → Gelato
Decentralized sequencing + AVS validators? → AltLayer
6. Risks & Limitations
RaaS is powerful, but comes with tradeoffs and risks that teams should understand before committing:
⚠️ Vendor Lock-in
Once your chain is live, switching providers is nearly impossible. Your bridge contracts, state, validator set, and ecosystem are all tied to your provider's infrastructure. Choose carefully.
⚠️ Centralized Sequencing (Most RaaS)
Conduit, Caldera, and Gelato default to centralized sequencers run by the provider. This is operationally simpler but creates a trust assumption. Only AltLayer decentralizes sequencing by default.
⚠️ Fragmented Liquidity
With 100+ rollups live, liquidity is fragmented. Your new rollup won't have deep DEX markets on day-one. This matters if you're competing with Optimism or Arbitrum.
⚠️ Provider Dependency on Upgrades
If your provider releases a critical fix, you're dependent on their deployment timeline. No provider has broken a chain, but the risk exists.
⚠️ Data Availability Concentration
If you use Ethereum blobs and they get congested, your DA costs spike. If you use Celestia and it experiences downtime, your rollup stalls. Diversifying DA is expensive.
⚠️ Governance Challenges
How do you upgrade your rollup if the provider wants to change the framework? This is an understudied problem. Community forks may be necessary.
Despite these risks, RaaS remains the best path to chain deployment in 2026. The alternative—building your own infrastructure—is even riskier. The key is informed decision-making: understand the risks, pick a provider aligned with your long-term vision, and plan for eventual decentralization (including sequencer decentralization via AVS).
Frequently Asked Questions
What exactly is Rollup-as-a-Service (RaaS)?
RaaS (Rollup-as-a-Service) is a managed platform that allows teams to deploy custom L2 or L3 rollup chains without building infrastructure from scratch. Instead of managing sequencers, handling data availability posting, maintaining bridges, and running RPC nodes, you use a RaaS provider's infrastructure. Think of it as AWS for blockchains—you focus on your application logic, and the provider handles the operational complexity. Popular RaaS platforms in 2026 include Conduit, Caldera, Gelato, and AltLayer.
How is RaaS different from a blockchain-as-a-service (BaaS)?
BaaS typically refers to managed infrastructure for existing blockchains (like Infura for Ethereum nodes), while RaaS specifically enables you to deploy your own custom rollup chain with your own execution environment. With RaaS, you get a unique blockchain with your own validator set and customizable parameters. BaaS is infrastructure rental; RaaS is chain deployment. RaaS is more powerful but requires more thought around tokenomics, governance, and security.
What's the difference between an OP Stack and Arbitrum Orbit rollup?
OP Stack (built by Optimism) and Arbitrum Orbit (Arbitrum's framework) are both frameworks for launching rollups, but they differ in design philosophy. OP Stack uses optimistic rollups (transactions assumed valid until proven invalid) and is modular—you can swap data availability layers, sequencers, and more. Arbitrum Orbit is based on Arbitrum's AnyTrust protocol and defaults to a more integrated stack. Caldera supports both plus ZK frameworks; Conduit specializes in OP Stack. Your choice affects cost, customization, and ecosystem compatibility.
Do I need to run my own sequencer with RaaS?
No—that's the entire point of RaaS. Most RaaS providers (Conduit, Caldera, Gelato) run centralized sequencers for you, which simplifies operations and gets you to mainnet faster. However, AltLayer's unique approach enables 'restaked rollups' with decentralized sequencing via AVS (Actively Validated Services), giving you more decentralization if you want it. For maximum decentralization but maximum complexity, you could run your own sequencer separately, but most new chains use provider-managed sequencers initially.
What are the typical costs of deploying a rollup via RaaS?
RaaS providers typically charge a combination of: (1) Deployment fees—usually $0 to a few thousand USD for setup, (2) Data availability costs—varies based on your DA choice (Ethereum blobs are cheapest, Celestia/Avail are moderate), (3) Sequencer fees—either a fixed monthly fee ($1k-$10k+) or a percentage of block space revenue, (4) RPC/infrastructure fees—per API call or bandwidth. Exact pricing depends on your throughput, data availability layer, and provider. Start with Conduit or Caldera's pricing pages for current rates.
Can I change my RaaS provider after launch?
Switching RaaS providers after mainnet launch is extremely difficult—it requires migrating state, adjusting bridge infrastructure, and coordinating node operators. This is the 'vendor lock-in' risk of RaaS. To mitigate: choose a provider with good track record, use modular stacks (like OP Stack) that don't lock you into a single tech, and keep your validator set diverse. Some providers (Caldera) emphasize flexibility and infrastructure portability, making future transitions easier than others.
Related Articles
⚠️ This guide is for informational purposes only. It is not financial advice. Deploying a rollup involves technical and business risks. Always perform security audits, review provider SLAs, and plan for long-term governance before launch. This guide reflects the state of RaaS providers as of March 2026 and may become outdated as the industry evolves.