Cardano Leios Models See MEV Risk for Only 2.6% of Analyzed Volume

Input Output Research don present new findings about the security and performance modeling for Ouroboros Leios, the Layer 1 scaling protocol wey Cardano propose. Research fellow Giorgos Panagiotakos share the results during one Cardano R&D session on July 16, 2026.

The presentation be part of the midyear update for WP1.1 Consensus, one research package wey cover Leios for higher throughput and Ouroboros Peras for faster transaction settlement. The work include network modeling under load, formal security analysis, voting certificate optimization and parameter selection for engineering implementation.

Leios Models See Small MEV Exposure When Network Dey Busy

The research team analyze how the higher transaction capacity wey dem plan for Leios fit affect mempool fragmentation, transaction ordering and front-running chances.

MEV, or maximal extractable value, na value wey person fit capture by changing the order, inclusion or exclusion of transactions before confirmation. Front-running na one type of MEV wey attacker go observe pending transaction and try put competing transaction ahead of am.

According to Panagiotakos’ presentation, about 2.6% of the volume wey dem examine for the research fit get MEV exploit. Front-running generally no dey profitable for the tested scenarios, and dem link the result to Cardano’s EUTXO architecture and how Leios distribute transaction processing.

The estimate apply to the models and conditions wey the research include. The public Post-CIP R&D findings support the broader conclusion say na only small fraction of transactions dey susceptible to front-running, even though dem no include the dataset or the calculation behind the 2.6% figure.

The public findings still talk say the size of the attack surface depend on the design of each decentralized application. Mempool-based attack go need control over relay infrastructure and ability to build competing transaction quick, while block producer chance to succeed go match the size of im stake.

Leios mempool simulations find say Cardano network topology get strong connectivity, e help rapid transaction diffusion and e limit fragmentation. When demand remain inside network capacity, the models report global mempool synchronization above 90% and fragmentation below 10%. When demand heavy pass capacity, diffusion fit slow and fragmentation go increase according to the load.

Markov and Resource Models Dey Guide Leios Parameters

Ouroboros Leios dey designed to boost Cardano throughput by allowing ranking blocks to include certified references to endorser blocks, wey represent bigger batches of transactions. Committees dey vote on those blocks before their certificates fit enter the ledger.

The research include one Markov model wey estimate the probability say endorser block go receive certification as ranking blocks dey produced. E track block production, certificate availability and whether dem gather enough votes before the next relevant block-forging opportunity.

These results dey help researchers evaluate Leios efficiency and choose protocol parameters based on modeled network behavior.

Another constraint model measure processor and bandwidth requirements. According to the published findings, to wait for transaction bodies and to apply transactions to the ledger na the two biggest sources of delay. Signature verification and Plutus execution no too restrict because many of that work fit run in parallel.

The models show say three CPU cores dey enough for typical 12 MB endorser block. If attacker release transaction bodies late on purpose, e fit double peak processor demand.

Those results dey inform limits like maximum number of transactions per endorser block, maximum block size and Plutus execution limits. The targeted throughput and hardware requirements fit therefore dey evaluated against measured resource behavior before dem select production parameters.

The security work still include modeling how endorser blocks go diffuse on time under adversarial conditions, formal proofs of safety and liveness properties for Agda and one trace verifier. The verifier dey compare implementation behavior with the formal protocol specification to find any deviation for generated blocks, certificates and ledger states.

The official Leios repository get the formal specification, simulations, network visualizers, cost analysis and trace verification tools. The project now join protocol research with active design and prototyping work for future Cardano node implementations.

New Voting Certificates Go Connect Leios and Peras

Input Output Research dey also develop optimized voting certificate design wey go work for both Leios and Peras. The goals na to reduce certificate size and voting traffic and still improve resistance against stronger adversarial models.

For Leios, certificates dey confirm say endorser blocks get the committee support wey dem require. For Peras, voting certificates dey support block boosting, one mechanism wey give extra weight to chain wey get enough endorsement and e speed up settlement.

The design build on earlier Input Output Research work, including Mithril’s stake-based threshold multisignatures and the Fait Accompli approach to committee selection. One research paper wey go describe the optimized certificates dey scheduled to complete before the end of 2026.

The research plan for the second half still include one technical report on components wey fit dey shared between Leios and Peras, plus separate analysis of timely endorser block diffusion under attack. Andre Knispel, Brian Bush, Matthias Fitzi, William Wolff, Yves Hauser, David Rosales and Sandro Coretti-Drayton be among the researchers wey dem list as contributors.

The three scheduled outputs go give engineering teams one proposed certificate format, documented links between Leios and Peras, and one security model for endorser block diffusion. Together, dem go define several parameters and adversarial assumptions wey needed for the next stage of both protocol implementations.