How Does Sui Work?
Sui is a Layer 1 blockchain built by Mysten Labs (founded by former Meta/Diem engineers) using the Move programming language and a novel object-centric data model designed for high-throughput parallel execution. Its C+ risk grade reflects the novelty of its untested consensus (Mysticeti) and execution model, significant validator centralization demonstrated during the May 2025 Cetus exploit recovery, and substantial token unlock overhang with 60%+ of supply still vesting through 2030.
TVL
$563M
Sector
L1
Risk Grade
C+
Value Grade
C+
Core Mechanisms
3.1.1
Delegated PoS with epoch-based staking rewards distributed pro-rata to validators and their delegators
Standard delegated PoS staking model. Validators earn rewards proportional to stake with commission rates. Delegators choose validators and share in epoch rewards. Epoch transitions every ~24 hours.
3.3.1
Direct delegation where SUI holders choose validators with no minimum stake requirement
Standard direct delegation pattern. Currently ~114 validators on mainnet. Stake concentration concerns exist with top validators controlling significant share.
1.2.1
Linear vesting with 1-year cliff for investors (14%) and early contributors (20%), unlocking over 3 years total
Standard VC-backed vesting schedule. Series A ($36M) and Series B ($300M from a16z, FTX Ventures, others). Cliff unlocks completed in 2024. Linear unlock continues through 2026-2027.
2.3.2
Sui Foundation manages 50%+ of SUI supply as Community Reserve with discretionary spending on grants, delegation programs, and ecosystem development
Foundation-managed treasury pattern. Sui Foundation controls the largest single allocation of SUI tokens. Centralization risk but standard for L1 chains. $10M security initiative launched September 2025 post-Cetus exploit.
2.1.4
NovelGas pricing with validator reference price voting — validators propose reference prices per epoch, with 2/3 stake-weighted quorum setting the gas price
Novel gas pricing mechanism. Validators vote on reference gas prices each epoch. Mechanism incentivizes validators to set honest prices via tallying rule that rewards those near the consensus price. Deviates from standard EIP-1559 auction model.
Novel/Storage-Fund
NovelRefundable storage deposit system — users pay storage fees when writing on-chain data, which are deposited into a fund that earns staking rewards; fees are refunded when data is deleted
Novel economic mechanism unique to Sui. The storage fund stakes its holdings and distributes rewards to validators for storage costs. Creates a perpetual funding source for storage without ongoing inflation. Refund mechanism incentivizes state cleanup.
Novel/Mysticeti-Consensus
NovelMysticeti v2 DAG-based BFT consensus requiring 3f+1 validators with 3-round commit latency, merging validation and consensus into single process for sub-second finality
Novel consensus design that replaced Narwhal/Bullshark. Uses directed acyclic graph for parallel block processing. Mysticeti v2 (November 2025) further optimized with unified validation and new transaction driver. Matches theoretical minimum for BFT round complexity.
Novel/Object-Centric-Execution
NovelObject-centric data model enabling parallel transaction execution on non-overlapping object sets without global state contention; native private transactions via ZKP added in 2026
Novel execution model fundamental to Sui's architecture. 2026 additions include native ZKP-based private transactions. Move VM 2.0 upgrade reduced gas fees by 40%. Differentiated from account-based models (Ethereum) and UTXO models (Bitcoin).
How the Pieces Interact
The May 2025 Cetus incident proved that a 2/3 supermajority of validators can coordinate to freeze specific addresses within hours. While used defensively, this same capability could be used to censor legitimate transactions. With only ~114 validators and significant stake concentration, the threshold for coordinated censorship is lower than for networks with thousands of validators.
Ongoing unlock of 6B+ SUI tokens through 2030 creates persistent sell pressure. April 2026 alone sees 42.9M tokens unlock. If DeFi TVL or network usage stalls while unlocks continue, token price decline reduces the real value of staking rewards, potentially triggering a validator economics death spiral.
The Sui Foundation controls 50%+ of token supply and operates delegation programs that direct stake to validators. This creates a single entity that can effectively control validator selection and network governance through economic incentives, despite the formally decentralized validator set.
The storage fund earns staking rewards that are distributed to validators. As the fund grows, it becomes a larger share of total stake, diluting rewards to external delegators. Conversely, mass data deletion could trigger large refunds, reducing the fund's stake and creating validator reward volatility.
Sui's parallel execution assumes transactions touch disjoint object sets. Complex DeFi transactions (flash loans, multi-hop swaps, liquidations) frequently require shared state, creating contention points that reduce parallelism benefits. The Cetus exploit demonstrated that single shared objects (AMM pools) can become critical bottleneck/attack points.
What Could Go Wrong
- Sui validators demonstrated the ability to freeze $162M in stolen funds within hours during the May 2025 Cetus exploit — a recovery success, but also proof that a coordinated supermajority of validators can censor arbitrary addresses, undermining the censorship-resistance claim.
- Over 6 billion SUI tokens remain locked in vesting schedules through 2030. With only ~3.8B tokens circulating, ongoing unlocks create persistent sell pressure equivalent to 60%+ of current supply still to enter the market.
- Sui's object-centric execution model and Mysticeti DAG consensus are novel designs with under 3 years of mainnet operation. Novel consensus and execution models carry higher risk of undiscovered edge-case failures compared to battle-tested alternatives.
- The Sui Foundation controls 50%+ of total token supply through the Community Reserve, and Mysten Labs holds an additional 10% treasury allocation, creating significant centralization of economic power in two related entities.
Validator Cartel Censorship and Network Confidence Collapse
TailTrigger: A regulatory authority pressures the Sui Foundation to blacklist specific addresses, or a coordinated group of validators holding >67% of stake decides to censor transactions for economic or political reasons, building on the precedent set by the May 2025 Cetus freeze
- 1.Validators holding >67% of stake implement address-level transaction filtering, either under regulatory pressure or coordinated economic self-interest — Targeted addresses cannot submit transactions on the Sui network; their assets are effectively frozen without any on-chain governance vote
- 2.DeFi protocols on Sui begin building censorship-resistance workarounds (e.g., private mempools, relay networks) — Network fractures into censored and uncensored transaction paths, undermining the uniform security model
- 3.Large DeFi protocols (Suilend, Bluefin) evaluate migration to other chains as users express concerns about asset freezeability — TVL begins flowing out as institutional and whale capital seeks chains without demonstrated censorship capability
- 4.Token price drops 30-50% as market prices in the censorship risk premium — Validator staking economics deteriorate; marginal validators exit, further concentrating stake among remaining validators and worsening the censorship dynamic
- 5.Foundation intervenes with emergency governance changes to limit validator censorship power — Foundation intervention itself demonstrates centralization, deepening confidence crisis regardless of the policy outcome
Risk Profile at a Glance
Overall: C+ (42/100)
Lower score = safer