Introduction

Union is a decentralized general message-passing interoperability chain designed to connect diverse blockchains across various runtimes and execution environments. In essence, a source chain sends a packet to Union, and Union relays it to the destination chain. The destination chain can trust that the packet is valid because of the consensus verification performed by the Union protocol’s validator set. For a more thorough introduction to Union’s architecture, we recommended reading our prior report here.

Union incorporates several technical innovations that significantly broaden and improve the performance of its offering: state lenses, zero-knowledge proofs, and intent-based filling, among others. These innovations allow Union’s interoperability protocol to achieve low latency, low cost, and broad functionality without compromising security or trust assumptions. In this report, we explain some of the key technical aspects of Union’s design and the improved performance that they unlock.

Union’s Architecture and Consensus Verification

Union operates its own blockchain as the hub for crosschain communication, leveraging CometBLS for consensus across a large set of validators. CometBLS offers a purpose-built consensus mechanism for Union’s interoperability use cases, improving upon the legacy CometBFT. This new mechanism is designed specifically to minimize the data, costs, and latency required for crosschain interactions. This design choice means that bridging with Union is based on consensus verification – consensus is achieved on a source chain’s state by Union’s validator set, rather than by a small set of custodians, and a proof of consensus is delivered to the destination chain.

Competing bridges like LayerZero and Circle’s CCTP use a Proof-of-Authority model (trusting specific off-chain signers or oracles), while others like Axelar or Wormhole use multi-party computation (MPC) with 10–20 guardians signing attestations. In contrast, Union “trusts the chain itself.” As long as a supermajority of Union’s validators are honest, the bridge is secure. 

Union’s consensus framework can functionally scale to a set of hundreds of validators, vastly outscaling the signer sets of MPC-based bridges.

State Lenses: Unlocking Crosschain State Access

One of Union’s core innovations is the concept of state lenses. A state lens is essentially “a conditional light client designed to extract state roots of specific chains from the state root of Union.” In simpler terms, Union’s blockchain maintains just enough information about connected chains, such as their latest state root or block headers, so that a destination chain can trust proofs of events from a source chain by checking Union’s state. This allows a destination chain to maintain a line of sight into a source chain’s state. This preserves the original packet data and sequence, enabling end-to-end traceability and more complex routing. It also means Union can carry any type of data (contract calls, governance votes, NFT messages, etc.), not just token transfers, because the mechanism is generic packet verification rather than token-specific. The introduction of state lenses addresses these issues by eliminating the need to rewrite packet data and by generalizing interoperability beyond purely token transfers. 

The prior method for routing transfers through IBC-connected chains used Packet-Forward Middleware (PFM), a module that forwarded IBC token transfers through the Union chain. PFM presented a number of drawbacks. It was limited to token transfers and couldn’t handle arbitrary data or messages. It would also alter the original packet data. PFM would attach routing info to the token transfer, then Union would strip that info and create a new transfer to the destination. This broke the continuity of the packet’s history as the original and forwarded packets had different proofs and sequence numbers, making onchain auditing and multi-hop routing difficult. In short, PFM could not feasibly scale beyond simple token transfers.

Source: Union Research

State lenses introduce a number of performance unlocks. They create what is effectively* a direct line of communication between connected chains, and eliminate *the need for costly and redundant re-execution of PFM transactions. Because Union’s state root encapsulates source chain events, third-party applications can use this data to anticipate incoming transfers and act on them quickly. State lenses can allow for faster crosschain interactions by exposing verifiable state earlier in the transaction lifecycle, as relayers and solvers can pre-fetch verifiable state from Union to enable faster message propagation and execution.

zkGASM: Gas Efficiency through Zero-Knowledge Proofs

Another edge in Union’s technical design is its gas efficiency for crosschain operations through the use of ZK cryptography, seeking to minimize gas costs in bridging by optimising the gas consumption at each step in the transaction lifecycle. A significant contributor to gas savings is the use of ZK proofs for light client updates. In Union’s protocol, when the Union chain needs to update its knowledge of a source chain’s state, it doesn’t rely on heavy onchain processing or multi-sig verification. Instead, it uses a ZK proof to prove the validity of the new state to the destination chain’s contract. Verifying a single ZK proof onchain can be highly gas-efficient when compared to verifying many individual signatures or storing entire block headers. A single client update on Union can batch and verify an arbitrary amount of transfers from an arbitrary number of chains, so the per-transfer cost approaches zero at scale. This design significantly reduces the per-message overhead and gas requirements.

Source: Union Research

In a benchmark evaluation of bridge protocols over 6 months, Union achieved the lowest gas cost per transfer among major interoperability solutions. An average crosschain transfer via Union used only about 138,377 gas in total (source and destination chain), whereas LayerZero required roughly 280,330 gas and Axelar about 848,548 gas for the same scenario. Wormhole’s transfers were around 392,789 gas, also far higher than Union’s. With 3–6x gas savings when compared to alternative interoperability providers, Union’s approach can be more cost-effective for users.

Source: Union Research

The biggest savings were observed on the destination side, as Union avoids expensive signature verifications on the destination. For example, Axelar’s high gas usage comes largely from verifying its validator threshold signatures on the destination chain, and Wormhole’s from posting guardian signatures. Union reduces these costs by batching multiple transfers together, contributing to the low per-message gas cost.

Union’s gas figures above considered the “critical path” of a transfer, which involve the steps needed to get the user’s funds delivered on the destination chain. Union’s design splits the transfer into a “critical phase” for delivering the packet and unlocking user funds, and an “acknowledgement phase” for post-delivery settlement. By optimizing the critical phase with ZK proofs and batching, Union ensures the fastest, cheapest user experience, while the secondary acknowledgement phase can happen asynchronously without affecting user wait time. 

Intents Bridging: The Fast Path through Intents

Aside from cost advantages, Union can lower latency for crosschain operations through an intent-based filling mechanism. With intents, an open marketplace of solvers can compete to fulfill transfers instantly. Normally, when bridging assets, users must wait for finality on the source chain and for confirmation of the message to propagate to the destination. This can present lengthy delays, particularly for certain networks that have long times to reach finality. Union’s standard mode does wait for source chain finality to protect against reorgs or double-spends. However, Union’s design offers a parallel path with intents. If a user expresses an intent, a third-party solver can step in and fulfill it immediately on the destination, without waiting for the source chain’s finality.

Solver settlement is coordinated through Union’s acknowledgment phase. The acknowledgement carries the information about which solver filled the transfer, so that the user’s escrowed funds on the source can be unlocked to pay back the solver. Essentially, solvers predict that a user’s transfer is likely to be finalized, and they provide liquidity upfront. By opting in to intents, users get near-instant transfers at the cost of a slightly higher fee, while solvers compete to earn those fees by taking on the finality risk.

Conclusion

Union’s technical design brings measured improvements to crosschain interoperability. By combining a consensus-verified hub with novel constructs like state lenses and ZK proofs for client updates, Union achieves an interoperability protocol that is highly performant, trust-minimized, and scalable. 

The use of state lenses resolves the legacy issues of packet forward middleware, enabling general message passing and multi-hop routing without data loss or costly re-execution. Additionally, the use of ZK proofs and other gas optimizations make Union one of the most gas-efficient bridges, as demonstrated by the benchmark comparisons. The introduction of intent-based filling adds a complimentary route, giving users the option of near-instant transfers filled by third-party solvers while maintaining trust-minimization. 

This research report has been funded by Union.fi Labs. By providing this disclosure, we aim to ensure that the research reported in this document is conducted with objectivity and transparency.  Blockworks Research makes the following disclosures: 1) Research Funding: The research reported in this document has been funded by Union.fi Labs. The sponsor may have input on the content of the report, but Blockworks Research maintains editorial control over the final report to retain data accuracy and objectivity. All published reports by Blockworks Research are reviewed by internal independent parties to prevent bias. 2) Researchers submit financial conflict of interest (FCOI) disclosures on a monthly basis that are reviewed by appropriate internal parties. Readers are advised to conduct their own independent research and seek the advice of a qualified financial advisor before making any investment decisions.