Leios Testnet Moves Cardano Scaling From Research Toward Real World Testing

Carlos Lopez De Lara, product manager for Leios at Input Output, appeared in the premiere episode of Block//45 to explain Cardano’s next major scaling track.

The interview focused on how Leios adds a new block structure to Cardano, why the Musashi Dojo testnet is being prepared, and what SPOs and dApp developers are expected to test before any future mainnet decision.

The interview featured Carlos Lopez De Lara, @carloslodelar, and was hosted by Lily Brodi, @lilybrodi, with Paul from Cardano With Paul, @CardanoWithPaul.

Leios Adds Endorser Blocks to Cardano’s Praos Model

Lopez De Lara described Leios as a Cardano scaling solution that uses the time between Ouroboros Praos blocks to create an additional block type known as an endorser block. Praos remains the base protocol and security layer, while Leios works as an overlay protocol designed to add throughput when network demand requires it.

Under the model he described, an SPO elected as slot leader can announce that it is creating an endorser block. The standard Praos block and the endorser block then move through the network in parallel. If the endorser block is validated, the next Praos block includes a certificate containing votes from SPOs that validated it.

The purpose of the endorser block is to carry more transaction data than a regular Praos block. Lopez De Lara said Cardano currently processes about 4.5 transaction kilobytes per second and that Leios could scale the system up to 200 transaction kilobytes per second. He also said the initial configuration is not expected to start at the upper limit, because a large immediate increase would create unnecessary costs for SPOs.

Instead, the early target could begin around two to five times current throughput, while the underlying machinery would be in place for later scaling. That approach positions Leios as a staged throughput upgrade rather than an immediate jump to maximum capacity.

Lopez De Lara also connected Leios to Cardano’s long term economic model. He said the network’s 2030 vision, including a target of 27 million monthly transactions, cannot be reached with the current protocol alone. He also said transaction fees become more important as reserves decline, making higher throughput relevant to long term SPO profitability and network sustainability.

Musashi Dojo Will Test Leios in Five Phases

The interview also introduced Musashi Dojo, the upcoming Leios testnet expected at the end of June. Lopez De Lara described it as an intermediate milestone between the original IO Research work, the Cardano Improvement Proposal process, the formal specification and a future mainnet implementation.

He said the Leios work has moved from the original research paper into an optimized protocol proposal, then into a formal specification and now into a prototype implementation. Musashi Dojo is designed to move that prototype from a lab environment into real world conditions, using distributed infrastructure operated by SPOs across different regions, hardware setups and network environments.

The testnet name draws from Miyamoto Musashi, the Japanese samurai associated with a two sword strategy. Lopez De Lara used that comparison to explain the Leios design. Praos remains the smaller block and the network heartbeat, while the Leios endorser block acts as the larger block used for different throughput conditions.

The Musashi Dojo structure follows the five chapters of The Book of Five Rings. The Earth phase focuses on basic protocol design validation. The Water phase explores protocol parameters and how the system behaves under different settings. The Fire phase introduces real world infrastructure conditions, including different operating systems, software versions, architectures, hardware and physical network limits.

The Wind phase is planned for adversarial testing. Lopez De Lara said Leios will be placed under harsh conditions to test how the protocol behaves when facing difficult or hostile scenarios. The final phase, Void, is designed for final validation before mainnet preparation, including tagging the node, preparing the release and beginning the hard fork initiation process if the evidence supports that path.

SPOs and dApp Developers Are Asked to Train Before Mainnet

Lopez De Lara described Musashi Dojo as a training camp where the prototype implementation can produce failures, feedback and improvements before production grade code is considered. He said community participation will influence the quality of the testnet results and the level of confidence available before any future mainnet step.

For SPOs, he said the operational setup should remain close to the current Cardano node model. The main additional requirement is a new set of BLS keys used for validating and voting on endorser blocks. He compared their use and registration process to the existing VRF key model.

The Leios website is intended to serve as the main entry point for participants. Lopez De Lara said it will provide information for running the Leios node, along with tutorials, a faucet for test ada and a faucet for test delegation. For experienced SPOs, the testnet is presented as a familiar Cardano testing environment with one new protocol function added to the operational flow.

He also directed the message to dApp developers. According to Lopez De Lara, application teams should not wait for mainnet before testing how their protocols behave under Leios. Developers can deploy on Musashi Dojo, test user experience under different throughput conditions and adapt their applications before a future mainnet upgrade.

The interview also covered the high confidence track running alongside the release candidate work. Lopez De Lara said the software implementation is one track, while the second track focuses on evidence needed for a mainnet decision. That includes protocol design testing, parameter testing, adversarial scenarios and threat model validation.

Those results would be used by SPOs, DReps and Constitutional Committee members when reviewing any future hard fork vote. Lopez De Lara said the team is working toward production readiness by the end of 2026, with November mentioned as a possible readiness point and December as a possible hard fork window if development, testing and community participation continue at the required pace. Musashi Dojo now gives Cardano operators and application teams a defined environment to test Leios before the protocol is placed in front of governance participants for any mainnet decision.