State of Stylus: Breaking Barriers on Arbitrum
A Castle Labs Research Report
A Brief Introduction to Stylus
Among increasing competition in the Layer 2 (L2) landscape, Arbitrum has chosen Stylus as a different path than its competitors, with a focus on supporting multiple programming languages (winking an eye at mainstream developers) and opening up for new use cases, previously unfeasible.
Stylus focuses on breaking down barriers to access Arbitrum, both in terms of programming language knowledge and computational limitations of EVM environments.
On the one hand, Stylus is Arbitrum’s talent unlock for developers, making development on Arbitrum easier. Instead of being constricted by Solidity and the EVM, devs can now leverage programming languages they already know.
By adding a WebAssembly (WASM) VM alongside the EVM, Arbitrum lets teams build smart contracts in Rust, C and C++, expanding Arbitrum’s reach from thousands of Solidity devs to millions of mainstream engineers.
The bet is simple: broaden the developer funnel and new application classes follow (heavier compute, richer libraries, AI-adjacent workflows).
On the other hand, Stylus not only simplifies developer onboarding but also expands the design space on Arbitrum, enabling heavier computations that were previously prohibitive on Solidity.
This report evaluates the post-launch impact of this bet, focusing on what Stylus changed, how the Stylus Sprint concentrated early momentum (demand, approvals, milestones), and which projects signal the direction of this initiative with tangible deployments.
The Stylus Unlock
Stylus turns Arbitrum into the most developer-accessible L2 by reducing the learning curve.
One of the main obstacles to bringing more devs onchain is their need to learn new blockchain-based programming languages, such as Solidity.
With the Stylus update, Arbitrum now houses a second Virtual Machine (VM), the WASM VM, alongside the EVM, meaning devs can deploy code in languages like Rust, C, and C++.. Imagine this as a translator that can read code in other programming languages supported and translate it, allowing it to be executed as we expect and interoperable with Solidity.
This leads to a major expansion both in terms of talent and design space:
Talent Unlock: teams can build on Arbitrum without having to retool their entire stack and skills around Solidity. A Rust or C-based project can now deploy on Arbitrum without having to re-write its code.
Design Unlock: WASM unlocks mature ecosystems and tooling and makes memory- or compute-intensive logic economically viable, enabling categories that are impractical in Solidity-only environments (advanced cryptography, simulation-heavy DeFi, onchain AI agents).
If you are not a developer, you might not realise just how significant this is.
In the world, there are about a few hundred thousand Solidity developers, but over 2.8m (as of 2023 Rust developers and over 13 million C/C++ programmers. Rust also consistently ranks first as the most “favourite” programming language by developers. Making it as easy as possible for them to code onchain is the biggest obstacle for mainstream developer onboarding.
While blockchain ecosystems might focus on bootcamps and hackathons, these developers would still need to learn a new blockchain-based programming language and logic. With Stylus, this is no more: developers can keep their preferred languages and toolchains, yet ship dApps that behave like EVM-native.
The second biggest unlock is purely operational, as it concerns the computing power and memory required to run computationally heavy contracts in Solidity, compared to much lighter efforts in Rust. In fact, opcodes written in Rust are much faster than on an EVM and use <1 gas making them better suited for AI workloads and memory-heavy contracts.
Developing with Stylus is so cheap that Arbitrum had to introduce a new unit of payment, which is called an “ink” (1 gas = 10,000 ink).
In turn, the deployment of Stylus translates into two main advantages for Arbitrum:
Distribution: Lowering the opportunity costs of learning Solidity for developers is a major advantage in distribution and onboarding.
Product: The additional WASM machine enables heavier computing and novel and more powerful case studies.
This approach highlights a different strategic approach compared to other L2s, based on different design choices. For instance, the Optimism OP Stack focuses on modularity and EVM efficiency, while Stylus opens up to new programming languages with a focus on expanding the talent pool of developers who can be onboarded to Arbitrum. zkSync and Scroll, on the other hand, are focusing on zkEVM equivalence, while Polygon is focusing on interoperability through the AggLayer.
Instead of purely focusing on infrastructural design choices, Stylus allow Arbitrum to tackle two problems at the same time: developer accessibility and computational limitations.
By unlocking Rust, C, and C++ alongside Solidity, Stylus turns Arbitrum into the most language-inclusive L2, combining Ethereum’s liquidity with Solana-grade compute. In turn, this translates into a more competitive builder funnel, a broader talent base, and opening up to new applications, which are more computing-intensive and could not previously be implemented on Solidity.
Making Web3 development seamless requires focusing on real problems. Tackling the developer funnel is a great step towards this direction.
As always, creating and deploying new tech is only a minor part of this process. The major hurdle of an ecosystem is following up on its product updates and ensuring deep integration by new and existing protocols.
To contextualise this, the next section focuses on how Arbitrum has bootstrapped real usage of Stylus with specific initiatives.
The Stylus Sprint
Two months post-launch (applications opened 11 Nov 2024), the DAO funded a year-long Stylus Sprint to incentivise development and real usage of Stylus, with the following timeline:
Projects could either apply to a free track or focus on the Request for Proposals (RFPs), part of a broader strategy brought out by Arbitrum, with a focus on:
Tooling & migration: to reduce the switching friction between Solidity and WASM, with Rust SDKs, debugging/contract-size management and more tools.
New languages & ecosystems: support AssemblyScript and Move, CosmWASM and Solana migration to grow the builder funnel.
AI and education: onchain AI templates and DevRel to seed new categories and educate the market.
Demand swamped the initial 1m ARB budget, with over 147 applications (89 for the Open Track, and 58 for the RFP track) requesting >$32m, forcing a March 2025 renewal request for an additional 4m ARB, bringing a total of 5m ARB at the disposal of the committee.
By the time of renewal, 17 projects were approved.
The pattern is clear: there is more qualified WASM-on-EVM demand than the initial budget anticipated, and the DAO is willing to double down on this vertical.
The extension led to a strong increase in proposals, with a total of 26 approved projects out of 149 applicants.
This also resulted in funding over 2m ARB paid and getting already about 50% of milestones already completed.
The average grant size of proposals is about 250k, compared to a smaller 138k average for rejected projects. This could mean that the scope and scale of projects could be a decisive factor in their acceptance or rejection.
The Power of Stylus: Practical Applications from the Stylus Sprint
Among the projects approved in the different tracks, we highlight:
Open Track:
Redstone
RedStone leveraged the Stylus sprint grant to launch Rust versions of its pull and push oracles, making them cheaper. This way, projects building on Stylus will be able to leverage Redstone oracles natively.
Enclave
Enclave is a “compute provider” specialised in confidential programming. As part of the Stylus Sprint, they received 450k in ARB to launch Encrypted Execution Environments (E3s). In practice, projects will be able to leverage private datasets for confidential operations without having to compromise on security and data privacy.
Human Passport
Human Passport is a secure identity provider, focused on a “proof-of-personhood” sybil-resistant ID. It received 115k ARB to integrate Stylus inside their “Passport XYZ”, lowering the gas costs for transactions and making it seamless to integrate for Arbitrum projects.
With Passport xyz it is possible to aggregate several ID proofs to improve sybil-resistance and guarantee higher quality engagement and participation. Stylus developers can use it to introduce a higher level of granularity of access to their applications, focused on high-quality actors.
RFP Track:
Stylus SDK (tooling)
Part of the RFP vision was to enable the creation of tooling to simplify the migration between Solidity and other non-EVM languages introduced with Stylus.
The Wakeup Labs team has received 200k ARB in funding to develop a solution that compiles AssemblyScript code into WebAssembly, enabling Stylus smart contract development in TypeScrip. This is a major unlock to onboard more JavaScript developers which are already familiar with the programming language.
Trail of Bits (debugging tooling)
Trail of Bits received 440k ARB to bring its existing Solidity-to-WASM compiler (Solang), currently working for Solana and Polkadot, at the availability of Stylus developers. This way, developers will be able to leverage existing EVM contracts on Arbitrum’s WASM with Stylus, without having to rewrite their Solidity code.
Angel (ARC, Playground, Ember)
AI was one of the main targets of the RFPs. Collectively, Angel received 220k ARB to leverage AI for:
AI-powered debugging
Automatic completion of smart contract process
AI-secured validation
The Angel collective is composed by ARC, Playground, and Ember:
ARC is a Rust-based AI agent framework launched on Solana, with a focus on seamless transactions and deeper integrations between agents and digital services.
Playground is a tool provider for Rust-based smart contracts
EmberAI focuses on allowing anyone to vibecode AI-powered Defi strategies
Through a combination of the open track and the more guided RFP track, these projects successfully populated an initial landscape of tooling, AI solutions, compilers and other tools that developers onboarded have at their disposal once building with Stylus.
Aside the ones highlighted within the context of the Stylus Sprint, other Arbitrum protocols have seen the value and flexibility of Stylus and used it for custom deployments.
The next section provides additional case studies beyond projects funded by the Stylus Sprint, thereby demonstrating its practicality in real-world usage.
Building with Stylus: Expanding Beyond the Stylus Sprint
This section expands on what’s possible to build with Stylus, bringing more examples from projects that did not participate in the Stylus Sprint.
Among those, we selected Renegade, Superposition and Fairblock.
Renegade Finance
Renegade has been one of the first examples of projects leveraging Stylus in production since October 2024. It is a DEX which focuses on bringing dark pools onchain to improve the execution of trades in decentralised environments.
The Renegade team adopted Stylus as a solution to handle heavy computation for verifying zero-knowledge proofs, reducing gas costs.
Stylus ended up being their final choice due to:
Cost Saving: Stylus represented the most cost-effective decision with settlement costs of about $0.30, at least two orders of magnitude lower than on Ethereum Mainnet, and even more than other L2S for specific benchmarks. WASM is a strong companion for an EVM, allowing the execution of libraries and other heavy operations and loops that would otherwise be too expensive in Solidity.
Seamless Integration: The entirety of the Renegade infrastructure was built in Rust and was seamlessly reused onchain thanks to Stylus.
Tools and Ecosystem: Arbitrum has a dynamic and scalable ecosystem, with well-developed infrastructure, tooling and protocols.
Superposition
Superposition is an AMM with a superior risk model, promising more efficient borrowing, while protecting user funds. As part of their roadmap, they introduced a yield mechanism within their Fluid USDC as an incentive alignment mechanism for developers.
Initially, this was built on Ethereum. However, with time, it became unsustainable. While Solana was considered an alternative due to its fast transactions, it still suffered from limited composability and a closed ecosystem that was not EVM-compatible.
On the other hand, choosing Stylus enabled Superposition to accommodate the technical aspects of the solution while ensuring widespread adoption.
With the flexibility of Stylus, they were able to create:
A complex dual-yield distribution model for their token, where rewards are split across the transaction initiation and also its counterparty.
A Payment for Order Flow Model (which can be better understood as comparable to the Timeboost implementation), which enables Superposition to capture a share of the MEV and actually turn it into rewards for users.
Additional Reward Layer: Superposition included further layers to its rewards, including a liquidity mining infrastructure and prediction markets, targeted to increase onchain liquidity and offering novel use cases to keep its users engaged.
Tooling and Composability: Tools like the Rust SDK support seamless integration with other libraries, complex testing and expressivity, while maintaining security, making backend integrations easier. In practice, they have enabled the Superposition team to test new rewards and incentives more quickly.
This example illustrates how Stylus implementation is all-encompassing, not only for onboarding new developers but also for enabling new use cases.
Fairblock
Fairblock is a protocol focused on ensuring transaction confidentiality and privacy protection for users. Information leakage is one of the main bottlenecks to making stablecoins ready for institutional and mainstream adoption.
To solve this, Fairblock focuses on confidential stablecoins and protected trading to:
Accelerate institutional adoption
Improve onchain privacy
Removing reliance on centralised infrastructure
Fairblock features a general-purpose "auction engine” called DeBid, which decentralises a process traditionally run by big centralised servers.
DeBid allows permissionless auctions through encrypted bids, which are decrypted only after the auction ends. These use multi-party computation (MPC) and threshold identity-based encryption (IBE), advanced cryptography curves and operations which could not be efficiently implemented if not using Rust and precompilers.
The key benefit of using Stylus is its ability to leverage a Rust-based execution environment in Arbitrum for computing heavier loads, eliminating the need to outsource decryption and execution to third parties. Keep in mind that the interoperability goes both ways, with applications using Stylus being interoperable with Solidity and seamlessly integrated into Arbitrum’s Defi ecosystem.
As part of this, Fairblock leveraged DeBid to launch Fairates, a fixed-rate lending platform which allows borrowers and lenders to interact without the need for a central authority (“auctioneer”)
Stylus enables Fairblock to approach its broader vision of integrating confidentiality into transactions, ensuring these solutions are both compliant and attractive to institutional investors, with a heightened focus on cross-border payments and private data marketplaces for AI.
From confidential transactions to permissionless auctions, these case studies strengthen Stylus' value proposition beyond what is most commonly referred to as its biggest value of onboarding more developers, opening up more practical areas of design with regard to:
Gas cost reduction
Reduction of Computational Overhead
Expansion of use cases
More devs.
More use cases.
More computing.
Less costs.
What’s next for Stylus?
Food for Thought and Future Outlook
By widening the developer funnel and raising the ceiling on computational capacity, Arbitrum is differentiating itself with its MultiVM support.
The case studies above showcase how Stylus is enabling new categories (AI, cryptography-heavy applications and DeFi primitives, etc.) which were previously unfeasible in EVM environments.
Each in its own way, these projects benefit from Stylus’s flexibility, lower costs, and security.
The significant number of applications in the Stylus Sprint highlights the attractiveness of this proposition. While the Stylus Sprint has served as a bootstrapping initiative, this initial push now needs to be complemented with other funnels, perhaps with specific and defined targets across different verticals. To boost Stylus penetration, a strong complementary push is needed at an ecosystem level, through robust Developer Relations (DevRel) efforts to create a Stylus-first builder culture on Arbitrum.
Learning from insights and iterating is also a key cornerstone of successful grant programs. The Stylus Sprint should also be followed with an analysis of retention, as well as on the existing developments by ecosystem projects to quantify new developers added, new commits etc., which can be attributed to Stylus, to evaluate its impact better.
In the meantime, we expect these initial deployments to evolve in a full-fledged set of migration tools to eliminate any friction between Solidity and WASM, while simultaneously accelerating the adoption of more languages.
Stylus is now close to the launch of v0.10.
Here’s what Offchain Labs Stylus Product Manager had to say about its importance:
“Stylus v0.10 will be a major unlock for Stylus developers, introducing first-class support for multi-contract projects. With this upgrade, developers can include multiple contracts within a single Stylus project and import other contracts directly as Rust crates. This will make it far easier to build on existing work, improve code reuse, and manage complexity at scale—helping to supercharge the Stylus ecosystem.”
We also had the opportunity to inquire about what is required to enhance the Stylus's impact and how the stylus sprint should be followed up with by builder-focused initiatives:
“Redstone recently recreated their oracle contracts with Stylus, making it possible for developers to swap out the Solidity version with just a single contract address change—and immediately benefit from lower gas costs due to the new implementation.
That’s the kind of impact we want to see more of: utilities that slot seamlessly into developer workflows, so you don’t always have to write in Stylus to take advantage of Stylus. Of course, there are still challenges, like ensuring smooth compatibility with popular tools such as Foundry. The good news is that projects like the ArbOS–Foundry integration from the Stylus Sprint are already tackling this.
Looking ahead, we’d love to see more reusable Stylus modules—small, computationally heavy pieces of logic that can be isolated and shared across many different use cases. These kinds of building blocks could dramatically expand what’s possible with Stylus, and we’re excited to see what the community comes up with next.”
These include cryptographic-heavy DeFi implementations, AI agentic workflows, and anything else that was previously unfeasible on an EVM environment, now made cheaper and more secure through Stylus.
Last but not least, we emphasise the importance of avoiding fragmentation across all tools. Strong templates, libraries and guidelines should be put forward to ensure consistency, while Arbitrum as a whole should ensure to avoid talent dilution across the EVM/WASM stack.