From launch, Across has approached bridging not only as a technical puzzle, but as a financial engineering one as well. The bridge has been live for almost one and a half years, and in that time has bridged over $1.5 billion in assets across 5 chains, has grown to a team of 20, and raised $10MM in funding from strategic investors.

Our emphasis on doing “more with less” has always been a part of each protocol upgrade, and that is why Across is the leader in capital efficiency today.

The most valuable decentralized applications of the cross-chain future will be the same as they are today: financial applications. They will require trustless, secure and efficient solutions to transfer value across blockchains. As bridge security attracts scrutiny, bridging technology is evolving to use liquidity networks that offer better security and UX properties by only handling canonical tokens.

However, liquidity network models suffer from imperfect capital efficiency and fragmented liquidity. Capital efficiency measures how effectively capital is deployed to service transfer volume, and is reflected in the balance between fees charged to end-users and yields paid to LPs. In working with only canonical tokens, liquidity networks require capital on all chains they connect, but their ability to manage and rebalance this capital is what drives capital efficiency. The protocol that manages funds most efficiently will be able to charge the lowest sustainable fees, attract the most users, and retain attractive LP yields.

On a separate axis, third party bridges exist as faster alternatives to the native “canonical” bridges of each chain, and must compete on speed to provide a seamless bridging experience. And the biggest determinant of speed is finality — ensuring source transactions have settled before taking action on the destination chain. Bridge designs that enable 3rd party entities, and importantly not the protocol itself, to abstract away and take finality risk will be able to provide the best possible experience to end-users.

While recent advancements in arbitrary message bridges and messaging protocols are a step forward for cross-chain usability, token bridging will remain essential for any permutation of the cross-chain future. The challenges of capital efficiency, fragmented liquidity and finality risk, must be solved — and Across will be the bridge to solve them. We are solely focused on building the best possible technology to transfer value across blockchains.

Although the exact origins of the following words are unknown (many attribute these words to the great Confucius), the product crossroads that we find ourselves at reminds us of the importance of this proverb:

Across is designed from first principles to:

I. Maximize Capital Efficiency via its novel interest rate lending model, protocol-level rebalancing and dynamic fund management capabilities.
II. Unify Liquidity via a single sided liquidity pool on mainnet
III. Absorb finality risk via its network of fast relayers

But, we aren’t resting on those facts.

Looking to the future, Across is doubling down on building the best cross-chain technology to transfer value. Our focus over the next few quarters is developing foundational technologies to expand to multiple EVM chains, build a new primitive to send custom messages alongside token transfers, extend our lead in capital efficiency, and innovate on DAO governance.

Across’ design is a hub and spoke model that uses destination chain canonical bridges for native rebalancing. The protocol requires two components to function:

However, this limits the chains that Across can support, as many destinations don’t have a trustless “canonical” message or token bridge implementation.

To overcome this restriction, we’ve been hard at work developing the most significant change to Across since launch: The Universal Bridging Adapter (UBA), which enables Across to connect to any EVM bridge.

Through Zero Knowledge messaging built on Succinct and a novel pricing mechanism (explained below) built on top of our existing optimistic verification system, Across will be able to support any EVM bridge as a destination, with Avalanche and BNB Chain as early targets, and many others later in the year.

With Succinct, Across is building on the frontier by embedding ZK messaging technology to trustlessly send unidirectional messages from mainnet to any EVM chain, in effect replacing the “canonical” messaging system required to send repayment instructions to destination pools.

Furthermore, in order to rebalance pools without a canonical token bridge, Across is developing a pricing mechanism to reward or penalize flows to particular pools based on the global state of pools at any point in time. This new pricing system will be financially engineered to ensure a robust and positive experience that Across users expect:

As an added benefit, the upgrades required for this functionality enable Across to easily support more tokens, with wstETH and MATIC as early potential targets.

The above is being built on top of Across’ existing design, enabling the protocol to combine a novel interest rate lending model, optimistic verification, ZK messaging and dynamic pricing to power the cross chain future.

Across is focusing on building the best financially engineered solution to transfer value across blockchains. We also recognize that the cross-chain experience is frustrating for users. They have to manage funds and gas on multiple chains. They have to know what applications exist on which chains, how to get there and what to do when they get there. They have to think about where their funds are.

The often proposed solution to this problem is cross-chain applications built on top of arbitrary messaging protocols (or middleware / hub chains). But, as we’ve seen many times, making cross-chain applications (bridges) work correctly is incredibly complex. Adding this complexity to every application is a recipe for repeating the same security and complexity issues that bridges have, in all other crypto applications. Adding to the complexity is the fact that the value of the messages are unbounded in arbitrary messaging protocols. This makes it impossible to provide economic incentives to ensure the system responsible for verifying messages is not corruptible.

We think a future where distinct application deployments sit on each chain is much more realistic — in fact it’s already happening. But this just exacerbates the frustrating experience we started with.

Across is taking a different approach to solving this problem. With a concept called Composable Bridging, Across is adding the ability to include messages with bridge transactions, allowing a user to bundle instructions for what to do with the funds that are being sent. The advantages to this are two-fold:

But it goes beyond just that sequence of actions. A dapp or wallet could manage the user’s funds across chains via a smart wallet or custodian contract and incentivize relayers to take cross-chain actions on behalf of the user. This would allow dapps to abstract complexities of the cross-chain world away from the user while also keeping them out of protocol contracts.

This is the start of a powerful new primitive that extends Across as a platform to solve the interoperability challenges of the cross-chain world. Stay tuned for more to come on our vision of the cross-chain future.

Across is the only leading bridge to employ an interest rate model, where idle LP assets are borrowed to facilitate near immediate transfers. Relayers provide upfront capital for user transfers then ultimately get reimbursed from the LP after a proof of valid relay has been verified by UMA Optimistic Oracle.

The optimistic verification process relies on a permissionless network of actors to propose and dispute proofs. As proposing is completely permissionless, a challenge period of 2 hours is set to allow for protocol actors enough time to dispute malicious proposals. When adding in the time for repayments to settle, there is a repayment every 3 hours, or 8 repayments per day.

This relatively long repayment window creates an inefficiency in which relayer capital isn’t optimally deployed, essentially capping the maximum fast transfer throughput of the bridge (per 24 hours) as the sum of relayer capital times 8 repayment periods.

If the repayment cycle is shortened, relayers need less capital to service the same amount of transfer volume, which is a net benefit to the capital efficiency of the protocol. But this comes with a risk — not having enough time to dispute a malicious proposal.

Across is solving this with a concept called Capital Recycling, where we will enable the DAO to elect Proposers responsible for submitting valid relay proofs, allowing the challenge period to be significantly reduced. And since number of relayer repayments is inversely related to the challenge duration, relayers will get repaid at a higher frequency, and be able to recycle their capital almost continuously, increasing the relayer’s, and the protocol’s, capital efficiency. As capital efficiency is the relationship between the fees charged to the end user, and the yields paid to LPs, any increase in capital efficiency is a net benefit to every participant in the network — by being able to lower fees without lowering yields, we can attract more users and retain LP capital.

This doesn’t change the trustlessness or decentralization properties of Across because any actor can still dispute proposals, preserving the fact that only 1 honest actor is needed to keep the system secure. A permissioned proposer does come with a minimal trade-off in censorship-resistance, but we believe this is mitigated as the DAO can replace the proposer if it suspects censoring behavior.

As a hub and spoke model, one of the unique and compelling advantages of Across is protocol level rebalancing — the protocol itself can move funds between the hub pool on mainnet and pools on destination chains via canonical bridges. It doesn’t need to rely on external actors to rebalance (and extract fees).

However, Across’ protocol-level fund management is susceptible to needlessly locking capital in the 7-day canonical bridges. To address this, we are upgrading each destination pool with a feature called Predictive Rebalancing that sets a target threshold based on historical user flows, thus enabling the protocol to selectively not rebalance with the expectation that future netting will occur. Increased netting reduces capital flowing through the 7-day canonical bridges, making that capital available to service transfers, thereby increasing the protocol’s capital efficiency.

Across is committed to decentralization. We are integrating decentralized governance via oSnap which empowers the DAO to execute the results of Snapshot votes on-chain independently. This is a departure from the standard of most DAOs, which use mutlsigs to carry out a DAO’s votes, which is both censorable and centralized.

Snapshot is very flexible, and it has been calibrated so ACX tokenholders can participate in governance with tokens staked as bridge liquidity, as LP tokens on DEXs, and even ACX tokens on other chains. If you have exposure, no matter where you are, you have a vote.

One standout feature of Across’ governance will be its delegated permissions system. One headache most projects in the Web3/DeFi space have felt is that a decentralized consensus mechanism leads to slow decision making. Across’ governance has been designed to strike a balance that allows the protocol to move quickly on matters that require expedience, but give the ultimate say to the DAO’s tokenholders. This is in alignment with Vitalik’s conception of “the bulldozer vs vetocracy.” The way this is accomplished is by having certain roles that can be delegated by the DAO. Specifically, the Financial Engineering committee, which makes decisions about protocol parameters, will be an elected role controlled by the DAO. This means the committee is accountable to the DAO without having to seek its consensus each time an adjustment needs to be made.

Capital recycling, explained above, is another example of this philosophy applied. It works by having a permissioned proposer, but that permission can be rescinded by the DAO. Again, this allows the protocol to move quickly without sacrificing decentralization.

The Across community seeks deeper ACX liquidity, and themselves have come together to create a solution that they call “community owned liquidity.” A series of Across community NFTs will be minted and sold to generate funds that will be paired with an ACX grant and added to an existing liquidity pool. Each NFT will represent an equal portion of that position, and NFT holders will earn pro rata rewards. Join the discussion on the Across forum.

The future of interoperability may send bridges behind the curtain, pulling asset transfers away from the eyes of the end user, so that they don’t even know they’re happening. Until then, we want to elevate your bridging experience so much that you won’t want behind-the-scenes bridging. Although we can’t give away too many details at this time, we are working on brand and UI enhancements that will make your “traveling” experience on Across less of a task and more of a sensory thrill.

Across takes you to other worlds, and we want it to feel that way. Our world-building begins with a special gift for our airdrop recipients who have staked and not claimed their rewards for 100 days. The drop will be on the week of March 21st. You can catch a first look at the Across lore by collecting the following NFT from Galxe, which celebrates the recent Arbitrum campaign. And finally, you’ll soon be able to proudly boast your Across staking status by claiming your dynamic badge from rep3. We won’t spoil the surprise just yet, so stay tuned!

Across’ raison d’être is token bridging.

From day one, this bridge has been optimized for capital efficiency, through its interest rate lending model, unified liquidity, dynamic fund management, and network of fast relayers.

Across’ lead in capital efficiency will widen even more once capital recycling has been turned on, which is the ability for the DAO to elect permissioned proposers to submit valid relay proofs and dramatically shorten the challenge period. The same is true for predictive rebalancing, which enables the protocol to selectively not rebalance with the expectation that future netting will occur.

To grow the reach of Across, the team is developing a universal adapter that enables the protocol to connect to any EVM bridge using zero-knowledge messaging built on Succinct. Across will also introduce composable bridging, which allows transfers to include instructions on the destination chain, creating a powerful new primitive for the cross-chain future.

Finally, Across has adopted an optimistic governance model that gives its 3300+ tokenholders direct control over the treasury and the protocol. Across is decentralized and unstoppable.

This journey has deep meaning to the Across team, and we are grateful to be taking it alongside so many talented community leaders. If you see the same opportunity that they do in an optimistic, capital optimized bridge to service the cross-chain future, then you have found your people.

Across Protocol is an intents-based interoperability protocol, capable of filling and settling cross-chain intents. It is made up of the Across Bridge, a powerfully efficient cross-chain transfer tool for end users, Across+, a chain abstraction tool that utilizes cross-chain bridge hooks to fulfill user intents and Across Settlement, a settlement layer for all cross-chain intent order flow. As the multichain economy continues to evolve, intents-based settlement is the key to solving interoperability and Across is at the core of its execution.

Website | Forum | Discord | Twitter | YouTube | Docs | Github