The Web3 storage sector has long been constrained by the binary paradigm of ‘high-cost cold storage’ and ‘low-efficiency hot storage.’ Veteran projects like Filecoin and Arweave remain entrenched in their niches, struggling to overcome the ‘security–cost–programmability’ dilemma. Backed by Mysten Labs’ technological expertise and a $140 million private funding round, Walrus entered the market with a $2 billion valuation. By systematically combining technological paradigm reconfiguration, ecosystem symbiosis, and business model upgrades, Walrus has challenged the established norms of the sector and redefined the core principles of decentralized storage.
Unlike projects that focus on incremental improvements, Walrus aims to build an entirely new storage ecosystem. It no longer treats storage as a simple data hosting service. Through deep integration with the Sui ecosystem, Walrus creates next-generation storage infrastructure that is programmable, low-cost, and highly adaptable.
Technological Paradigm Reconstruction
Traditional storage protocols have long been trapped in a one-dimensional competition among security, cost, and efficiency. Filecoin ensures high security through 25x redundancy, inflating costs; Arweave sacrifices interactivity for permanence. Walrus breaks this cycle through its RedStuff two-dimensional erasure coding and programmable architecture, achieving a dynamic balance across all three dimensions. RedStuff splits and verifies data in a row-column matrix, ensuring 99.98% availability with only 4–5x redundancy. This dramatically reduces data loss probability compared to Filecoin and lowers annual storage costs for 100GB AI training data from $12,000 on Filecoin to $2,400. Costs are also about one-fifth of centralized cloud providers like AWS.
Walrus also introduces programmability through tight coupling with the Sui ecosystem. Stored data becomes dynamic, programmable objects on-chain. Developers can update NFT metadata in real time, manage hierarchical access for AI training data, and track RWA asset proof documents via Move smart contracts. During testnet trials, Decrypt Media automated copyright sharing for a 4K video library, improving efficiency by 30x over traditional methods. Technical optimizations further reduce data recovery time for high-frequency AI access by 40% compared to Arweave. Compliance needs are addressed through a global node alliance using zero-knowledge proofs, compatible with major regulations like GDPR and CCPA. However, relying on the Sui network for core functions introduces some latency when TPS exceeds 10,000, highlighting a potential autonomy limitation.
Ecosystem Symbiosis
Most Web3 storage projects rely superficially on public chains, but Walrus establishes a ‘mutual empowerment’ loop. Sui provides Walrus with a high-performance control plane for node management and economic incentives, allowing the team to focus on storage optimization. In return, Walrus becomes the native storage infrastructure for AI and RWA projects in the Sui ecosystem, creating synergy that attracts 14 million testnet accounts and processes 5 million data chunks, totaling 27.85TB of active storage.
Walrus also supports ecosystem growth financially, allocating 35% of funding to assist Sui developers and subsidize storage costs, activating internal demand and promoting adoption by 78% of ecosystem partners. Storage activities consume SUI tokens as gas, forming a positive loop: storage demand → SUI deflation → ecosystem value growth. The team is also developing cross-ecosystem interfaces for Ethereum and BSC, aiming to expand business share beyond Sui over time. However, most revenue (90%) still comes from Sui, and cross-ecosystem adaptation remains challenging.
Business Paradigm Upgrade
Traditional storage projects rely on capacity-based pricing, but Walrus builds an integrated business model: scenario-based pricing, token empowerment, and value-added services. AI storage follows a three-tier pricing system, including dynamic fees based on access frequency, computation coordination, and value-added services like copyright management. By October 2025, AI scenarios had stored 12TB of data, generating 42% of core revenue.
For RWA projects, tiered on-chain review fees, staking requirements, and long-term storage models generate stable revenue. For example, a commercial office RWA project in Shanghai earned nearly $200,000 through compliant review, storage, and traceability services, with a service cost of only 25%, making RWA the largest profit segment (47% of revenue).
Walrus uses a dual-token system (WAL and SUI) to capture value. WAL serves as payment, staking, and governance token, with buybacks and subsidies linking commercial gains to token appreciation. SUI is used for gas fees, lowering user barriers and forming a positive feedback loop between profitability, token value, and ecosystem expansion. Limitations remain, as AI and RWA account for over 90% of use cases, and deep engagement with top-tier clients is still limited.
Paradigm Impact and Industry Insights
Walrus’s innovations have reshaped its competitiveness and influenced the entire Web3 storage sector. Technologically, the RedStuff coding demonstrates that low cost, high security, and programmability can coexist. Ecosystem-wise, the mutual empowerment model sets a benchmark for deep chain–storage integration. Commercially, scenario-focused, value-added pricing proves that storage projects can compete on value rather than scale.
Challenges remain: dependence on the Sui ecosystem limits autonomy, cross-chain adaptation is difficult, and client concentration poses risk. Expanding into NFT storage, government data archiving, and diversifying clients will be key to sustaining growth.
Conclusion
Walrus’s success reflects a new era of paradigm-driven innovation in Web3 storage. By addressing high costs, low efficiency, and limited functionality, it has created a storage ecosystem valued at $2 billion. The future will test the team’s ability to maintain autonomy, expand cross-ecosystem, and diversify scenarios. If successful, Walrus could become a benchmark for the entire Web3 storage industry, proving that the next phase of competition is defined not by incremental metrics, but by the ability to innovate across technology, ecosystem, and business paradigms.