The roadmap, authored by IO Researcher Giorgos Panagiotakos, lays out a path to what IOG describes as a Layer 1 resolution of the blockchain trilemma: the longstanding engineering tension between scalability, decentralization, and security. A public testnet is targeted for June 2026, with mainnet activation contingent on a community governance vote under Cardano's Voltaire framework.

---

How Leios Works

Leios restructures how Cardano produces and validates blocks by splitting the process into three distinct layers. Input blocks, created roughly every five seconds, carry user transactions. Endorsement blocks bundle and verify those inputs through a stake-weighted lottery committee.

Ranking blocks, produced every 20 seconds using existing Praos mechanics, provide the final canonical ordering and security anchoring. The design allows hundreds of blocks to proceed in parallel within each 20-second window without compromising the security guarantees of the current protocol.

Three cryptographic tools support the architecture. SNARKs (Succinct Non-Interactive Arguments of Knowledge) let consumer-hardware nodes verify chain state without downloading the full blockchain. This capability is being built out through Mithril, Cardano's existing stake-based threshold signature scheme, which is being extended with recursive SNARK circuits; prototype testing was completed as of May 2026. Data Availability Sampling allows nodes to confirm that transaction data exists without retrieving all of it. Bloom filters reduce the volume of data nodes must process when participating in chain verification.

Together, these are intended to keep node operation within reach of consumer hardware, even as throughput scales. Post-Leios minimum node requirements are specified as 6 or more processor cores, a connection of 100 Mbps or faster, and solid-state storage.

UTXO sharding, also part of the plan, enables horizontal scaling by distributing state management across nodes and eliminating transaction conflicts. Cardano's Extended UTXO model is architecturally compatible with this approach because transactions declare their inputs and outputs upfront, making them independently verifiable and conflict-free by design.

---

The Numbers Behind the Upgrade

Cardano currently averages 0.73 transactions per second over one-hour windows, with a maximum observed rate of 11.62 TPS across any 100-block sample, according to Chainspect data. The theoretical ceiling under the present protocol sits at 18.02 TPS. Data throughput runs at roughly 4.4 kilobytes per second.

Post-Leios projections, governed by CIP-0164 (Linear Leios), put the target range at 200 to 1,500 TPS on mainnet, with a combined ceiling of up to 1 million TPS when Layer 2 infrastructure and EUTXO parallelism are included. Data throughput per peer would rise to approximately 900 kilobytes per second. Plutus smart contract execution limits are projected to increase roughly 100-fold, making application logic that is currently impractical viable on-chain.

The upgrade also pairs with Ouroboros Peras (CIP-0140), a separate protocol change targeting finality. Under the current system, high-confidence settlement can take hours. Peras introduces periodic voting rounds among block producers, compressing that window to approximately two minutes.

---

Why This Matters Outside the United States

Cardano's most active deployment region for practical use cases is Africa, where the network underpins land registry systems, carbon credit markets, rural energy payments, and agricultural traceability projects across more than a dozen countries. A March 2026 grant program funding $30 million across 14 African nations through Cardano's on-chain treasury drew 180 project submissions in its first week. The region's institutional footprint has grown alongside its developer activity: the Cardano Africa Tech Summit convened in Nairobi in February 2026, and African delegates represented one in five participants at the Cardano Constitutional Convention.

Current throughput constraints are not abstract for builders in these markets. A land rights platform operating in Mozambique and Ghana is migrating more than 80,000 records on-chain. A carbon registry in Tanzania is targeting 50,000 credits in its first year. A solar mini-grid payment system in Nigeria runs transactions that are highly sensitive to both confirmation speed and fee levels.

For each of these, the gap between Cardano's current 0.73 TPS average and a 2-minute finality target is a meaningful one. Several of these projects are already operational in early forms; the upgrade is widely seen as the prerequisite for moving from experimental deployments to fully scaled, commercially competitive products.

Alex Maaza, Sustainability and Innovation Lead at the Cardano Foundation, framed the broader stakes in January 2026: "We don't need another platform built elsewhere and exported here. We need infrastructure that empowers African builders to solve African problems on their own terms."

South Asia is also a relevant market. India and Pakistan recorded more than 40 percent year-over-year user growth on Cardano in 2025. The Gulf remittance corridor, one of the highest-volume in the world, stands to benefit from Peras finality that competes directly with services like SWIFT GPI. One infrastructure caveat applies: post-Leios nodes require connections of 100 Mbps or faster, a threshold that remains a practical constraint in lower-connectivity regions of South Asia where network infrastructure lags behind demand.

---

What Comes Next

IOG Chief Scientist Aggelos Kiayias has framed the longer-term goal in terms of resilience as much as speed: "The goal is to design for robust optimism. Acceleration should be robust so adversaries cannot easily disrupt the protocol to leave the optimistic path."

Beyond Leios, IOG is developing Ouroboros Omega, described as the protocol's intended final form. It introduces optimistic multi-path execution, where the network achieves near-centralized performance under favorable conditions and reverts to conservative decentralized operation when conditions deteriorate. A second design principle, Multi-Resource Security, diversifies the protocol's trust base beyond pure proof-of-stake by incorporating useful computation and reputation signals tied to digital identity.

Developers should note that wallets, indexers, block explorers, and APIs will require updates to handle the new block types before Leios goes live. Mainnet activation requires a governance vote, meaning African and South Asian DReps and stake pool operators have a formal role in the timeline. The underlying research was presented at Crypto '25 in August 2025 by Panagiotakos, Kiayias, and Alexander Russell, giving the proposal a peer-reviewed foundation before it reaches a governance vote.