A New Major Product Idea for the Cosmos Hub

[Noam]

Introducing: Halo, an ATOM-powered Shared Sequencer & Settlement Layer

How do we start driving some of the Cosmos ecosystem’s value back to ATOM? And how can we amplify the Cosmos Hub’s utility and power as a secure deployment and bridging platform?

Introducing HALO, a new project being researched and developed by a small team between Binary Builders & Informal Systems. A product we think could solve a few critical issues in the Cosmos ecosystem.

We believe this new product should be part of Cosmos Hub’s roadmap and I’d like to take some time to explain why, and open a discussion with the community about its place in the Hub’s future.

The AppChain Thesis and its Flaws

While Cosmos is known for pioneering the AppChain thesis where every protocol would have its own layer 1 blockchain, this has led to some obstacles that specifically hurt early-stage projects:

• AppChains are expensive: The average inflation for a Cosmos chain currently sits between 8-10%. Most of this inflation is directed towards delegators and validators. If you’re a super early stage layer 1 with a market cap of $20M, you’re printing 2 Million USD each year that’s hitting the open market, while likely producing very little transactions and sometimes even empty blocks. As we all know, this really hurts early-stage token economies.
• AppChains infra-operations are complex: Running and maintaining an active validator set is tricky to set up and manage. Community management and elaborate testnets take up valuable resources that slow down a project’s time-to-market, which results in the Interchain Stack becoming a less attractive starting point.
• AppChains increase fragmentation and can lead to sub-par UX: This is not just an issue in Cosmos, we’re seeing the same wheel being re-invented in the EVM L2 space as well, with users searching for the best ways to bridge tokens and sometimes getting stuck on other chains if they don’t have the right token or wallet available.

We believe the Cosmos Hub is well-positioned to address this problem and it would fit quite nicely within the Hub’s narrative of being a secure deployment & bridging Hub.

What is Halo?

Halo is a consumer chain that runs a decentralized shared sequencer to provide censorship resistance and liveness guarantees to rollups. It enables highly performant and cost-effective decentralized applications that can be built using the EVM, CosmWasm, or the highly configurable and powerful Cosmos-SDK.

The network also functions as a settlement layer that enables rollups to seamlessly connect to all the leading EVM networks, Solana, and every Cosmos chain & rollup using IBC.

Halo enables Cosmos chains to massively reduce operational costs and in some cases even completely eliminate inflation. Rollups pay on a per-transaction basis, which in many cases reduces chain operational costs by as much as 90%.

Features:

• 0.5-1.5 second blocktimes
• Pay-per-Transaction, no staking rewards required
• Free to use pricing oracle
• Instant bridging access to Cosmos, most EVM chains and rollups, Solana & more using one-click transactions from any network enabled through the Skip API
• No relayer required, rollup nodes automatically communicate IBC transactions with Halo

While research is still early, I’d like to walk you through the current thinking in terms of how Halo would work:

How it works - Sequencing

This is best explained by going over each step from the user starting a transaction to the rollup executing it.

1. A user submits a transaction via a website or directly on the wallet
2. The TX is sent to a rollup node
3. The rollup node performs a check to see if the TX is correct and the user has enough balance
4. The TX is then sent to Halo by the rollup node
5. Halo orders all the rollup’s TXs and puts them in a block, separated by each rollup’s namespace
6. In case we leverage Celestia, Halo would send the transactions to Celestia’s DA Layer
7. Rollups read from Halo, and optionally Celestia, to collect all the blocks, after which they’ll execute the transactions and change the state on the rollup (e.g. your balance)

Now, rollups can decide whether they want to settle on Halo, or perhaps remain sovereign or settle on a different network if they’d like.

How it works - Settlement & Bridging

Halo is a lazy sequencer in the sense that it has no awareness of what the rollup transactions do. It has no access to the state transition functions of any rollup, which enables it to be so performant. However, when it comes to settlement and the finalization of blocks, we need some kind of proving system to make sure the rollups behave correctly.

For those new to how rollups work I’ll try to TL;DR as much as I can: rollup nodes are tasked with executing the transactions, similar to validators, but there is no stake used in this system to prevent misbehavior. A rollup node could just lie about the state if it wants to. In order to trust the rollup node, we need some kind of proof.

Usually we see either Optimistic or Zero Knowledge (ZK) proofs being discussed. But in reality most rollups you use, like Base, still don’t have fully functional proving systems.

We aim to use ZK technology in the long term to offer rollups the most secure path forward, but the Interchain Stack is unfortunately not quite ready to be provable, nor are ZK proofs currently efficient enough to be used for this purpose.

Learning from Optimistic Rollups

Optimistic rollups have a waiting period of several days so that faults can be proven and shared. Theoretically a rollup would undo a transaction once the proof is shared. In practice we actually see third party node providers verifying whether or not the state is correct.

This is how you can avoid the waiting period and have instant withdrawals: some third party essentially sees your bridging request and says “I’ll check the transactions and execute the block myself, if it looks good I’ll front the money for the user and wait until the waiting period is over, for a small fee”.

This is exactly how we want to do settlement on Halo in the near term. We want to employ optional attestors, which are basically just rollup nodes that post the same state root to Halo as the “official” rollup node. Once enough attestations come in, we can consider the block final and allow the rollup to bridge to other networks.

Let’s look at another transaction lifecycle. If we continue from point 7 under sequencing:

8. Once a rollup node takes the blocks from Halo and execute all the transactions, a state root is posted to Halo, which is derived from the current state of the rollup’s node
9. Attestors, which are essentially whitelisted or elected node operators, are tasked with posting the exact same state root to Halo
10. Halo knows how many state roots are required to consider a block final. The rollup decides this at registration and it can be modified through whichever way the rollup wants to (e.g. governance).
11. Once enough matching state roots come in, Halo marks the block as finalized and we can consider the transaction completed. At this point bridging transactions are forwarded to other networks.

This essentially formalizes how Optimistic rollup networks currently work in practice. When ZK tech is ready, we migrate to that system.

Doesn’t this compete with Interchain Security?

Not at all, and it’s best to try to map out where Halo sits between ICS & smart contract deployments on consumer chains like Neutron and Evmos. See this overview:

Halo targets teams that want the performance and cost-effectiveness of rollups, but want to retain the power and customizability of the Cosmos SDK & IBC. There are some limitations to rollups such as the fact that CometBFT and ABCI’s voting extensions aren’t available to them, as well as the different security properties of running an L2. Projects that require more security and customization are better served by ICS.

What Halo essentially offers is a low-cost middle path between launching a smart contract and a full-blown layer 1 blockchain. We expect many projects to “upgrade” as they grow and we plan to offer very clean migration paths to our users.

How does it drive value?

Halo is a network that uses ATOM as its base token for governance, sequencing & settlement. Rollups are obviously free to use their own token on their network, but fees for sequencing, settlement and bridging are handled in ATOM.

For the price of running one consumer chain, we can compress the many, many Cosmos chains into a unified rollup ecosystem. It’s a win-win situation where Cosmos chains save a ton of cash while they acquire and spend some ATOM for the services provided by Halo.

EVM-alignment: What is this about using IBC to connect to all EVM networks?

In my work leading the Business Development team for the Interchain Stack, we have a lot of conversations with teams that are interested in Cosmos. Unfortunately, some end up going to Optimism, Polygon or other competitors. There are various reasons for this, but the most common one is that they want access to EVM users & liquidity.

I believe that neither Halo, nor ICS, is able to successfully compete as long as we cannot make that user-onboarding experience from EVM as good, or even better, than it currently is on EVM-aligned platforms.

As some of you might know, the IBC team is working on a much lighter version of IBC called IBC Eureka, which would make building an implementation for any new chain a lot simpler. Part of the rollout includes an IBC-Solidity contract that could be deployed on all major EVM L1s and L2s.

In my opinion it is critical that the Hub deploys these contracts and builds out the infrastructure required to maintain these bridges, finally bringing the Cosmos Hub bridging thesis to completion. We need every rollup and ICS chain to get free and truly permissionless access to all networks, without integration fees or a dependence on third-party bridging networks like Axelar that impose additional security risks to our users & the value that’s being transferred throughout our ecosystem.

Another key step here would be finding better EVM wallet support for Cosmos chains so that users coming from EVM-based networks are never stuck when they arrive in Cosmos, similar to how Osmosis recently approved Lite EVM. A partnership with the EvmOS team seems like a logical step to take here and we are already in talks with them about how to offer this to our users.

Learning From the Past: Going to Market

As we all know, just building good tech doesn’t get the job done. Fortunately, we have a solid 4-person business development team over at Binary Builders that’s currently doing work for the ICF and the Interchain Stack. Part of the work we do is identifying targets and reaching out to new leads about adopting the Cosmos SDK or integrating IBC. We’ve seen some good progress this year with Cardano, Thorchain, Near and several yet-unannounced projects integrating IBC, as well as the many Cosmos chains that are currently in the Interchain Builders Program.

Besides outreach, we also have a unique inbound advantage where we’re able to speak to the many early-stage / stealth-mode projects currently on the Builders Program, as well as leverage the funnel through the cosmos.network website.

A lot of this work is honestly just jumping on calls with teams, meeting them in person, and making a case for Cosmos while listening to their needs. Most of the time it’s not just a technical decision, but a business one. Who can you partner with in Cosmos? How do you access users & liquidity on major EVM networks? How can we support you in going to market? Fortunately we do have answers to most of these questions now.

We’re in talks with AADAO & Informal about how we can build out a comprehensive go-to-market strategy for Halo, as well as ICS V2. Some key things we want to do are:

• Extending Binary’s business development pipeline to include the Cosmos Hub
• Doing keynote presentations at key events like Cosmoverse as well as non-Cosmos events
• Organizing / attending Hackathons
• Setting up ATOM Economic Zone & Cosmos Events
• Establishing a small grants program to incentivize quality teams to build out MVPs in 2-3 month periods
• Enlisting a PR & Marketing agency to get wider exposure
• Better developer experiences (Cosmos Hub docs & tutorials are notoriously weak right now)

We want to make sure that the success of our rollup partners becomes our key priority. If you deploy on Halo, we can likely support you through the Builders Program, give you discounts on many service providers like auditors & market makers, free cloud credits from our partner providers, access to our VC network, as well as deep technical support from the core experts of the Interchain Stack. Co-marketing opportunities can obviously be explored as well.

I want to emphasize that we take marketing and business development extremely seriously and that we recognize that no effort should be wasted without a thorough go-to-market strategy & team.

There has been some coordination and alignment work done in the background between AADAO, Informal, Binary, Hypha & CryptoCrew, with the intent of addressing this specific issue. In the next few months I’m looking forward to sharing more about this.

Current Status & Timeline

We are currently working on the specification and getting into the nitty-gritty of how each component should effectively function, and we’re talking to teams, new and old, about their needs in order to learn what needs to be adjusted on the product side.

We have submitted a grant proposal to the AADAO to fund this work until the end of year, on which we’re currently waiting for an answer. Binary Builders are in talks with Informal Systems about doing a joint funding proposal that would ideally include Halo, as well as ICS and Gaia maintenance.

Depending on how funding develops and whether or not we will receive a grant, the roadmap might need to be adjusted.

What’s Next?

I’d love to hear from the community whether or not this project is of interest to ATOM holders. While many people I spoke to have told me to launch a token, I honestly believe it would be in the better interest of the Cosmos community as a whole to launch this as a purely ATOM-powered network so that all the future value can drive back to the Cosmos Hub and its community instead of anyone’s personal pockets.

I think what we’re doing here is a really good first step in modernizing the Cosmos ecosystem and making the ATOM Economic Zone a more attractive place to operate within, and I’d love to hear from the community whether or not we’re aligned on that vision If it’s looking like a yes, we’ll move this to a funding proposal towards the end of the year.

And lastly, if you’re a project that’s looking to build a highly performant, cost-effective and extremely configurable decentralized application, we’d love to chat! Please reach out to me on Twitter or via email, and I’d be happy to jump on a call to discuss how Halo can support your team.

[Govmos]

Thank you, @Noam, for this thorough and high-quality work. Binary Builders has once again delivered ideas and content that we take great pleasure in reviewing. We will share our feedback using our standard three-step process: context, analysis, and conclusions.

Context:

The concept of funding a decentralized sequencer deployed as a consumer chain leveraging the Cosmos Hub’s security is certainly worth exploring. Whether the product will achieve market fit is a separate question, which we will address by outlining the pros and cons in our review. Nevertheless, this discussion should primarily focus on “how” and “when” rather than "if." To address these critical questions, we will analyze the design choices and conclude with strategic recommendations aimed at maximizing the likelihood of successful market adoption.

Analysis:

Let’s begin by examining the proposed design choices, as quoted from the author:

@Noam

This is how you can avoid the waiting period and have instant withdrawals: some third party essentially sees your bridging request and says, “I’ll check the transactions and execute the block myself. If it looks good, I’ll front the money for the user and wait until the waiting period is over, for a small fee.”

We completely agree with this statement and have long been critical of the inherent inefficiencies in Optimistic systems. Their primary strength lies in operating at marginal costs, with relatively low technical overhead, allowing operators to process a high volume of inexpensive transactions. However, the trade-off comes in terms of security and centralization. Optimistic rollups often require external Data Availability (DA) layers, introducing third-party dependency risks and security assumptions. Many rollups mitigate these risks with extended withdrawal (or challenge) periods.

The proposal to have a third party front the money and wait out the withdrawal period is a sound economic solution but still introduces third-party reliance. It also necessitates centralized sequencing compromises to maintain low operating costs. Striking the right balance between these factors is crucial. We believe that introducing enforcement mechanisms, such as protocol-level censorship detection or subjective proof submission, could significantly benefit Halo. Leveraging the ICS slashing mechanism on the Hub would incentivize validators to ensure neutral and efficient sequencing. This is an area ripe for innovation, with the Hub’s instant finality slashing feature offering a solution that current EVM-based rollups can’t match, unless Ethereum successfully implements Single-Slot Finality (SSF). A feature that is highlighted in the Ethereum roadmap, but still in an early design stage.

@Noam

This is exactly how we want to do settlement on Halo in the near term. We want to employ optional attestors, which are essentially rollup nodes that post the same state root to Halo as the “official” rollup node. Once enough attestations come in, we can consider the block final and allow the rollup to bridge to other networks.

Halo’s instant finality slashing should be used to incentivize rapid attestations. We envision a system where revenue is distributed based on how quickly valid attestations are provided. Incorrect submissions would be slashed on the Hub, creating a balanced short-term risk-reward system for validators to sequence efficiently. This mechanism should be combined with a long-term protection system to mitigate risks of censorship and MEV.

@Noam

This essentially formalizes how Optimistic rollup networks currently work in practice. When ZK tech is ready, we migrate to that system.

Ethereum researchers have expressed a preference for ZK scaling, and we at Govmos fully agree. ZK technology, combined with the instant finality properties of Cosmos’ consensus, should be a key long-term consideration. Design choices should account for this future migration to avoid technical debt. We would appreciate Binary Builders’ input on developing a roadmap that minimizes this debt, optimizing longer-term migration efficiency.

Conclusions:

Our review highlights key strategic considerations. The product must be innovative and competitive, especially in the crowded rollup ecosystem where constant innovation makes it difficult to stay current. We believe the Cosmos Hub’s instant finality and slashing consensus mechanism is a powerful asset that can give Halo a competitive edge. However, Ethereum’s roadmap, which includes similar future developments, makes timing critical for Halo’s success.

We believe that the proposed Q3 2025 mainnet launch should be expedited to Q2 2025 to secure an advantageous market position. We look forward to hearing from the technical team to determine whether this timeline is feasible and whether additional resources might help accelerate the delivery process, thereby increasing the likelihood of success.

@Noam

What Halo essentially offers is a low-cost middle path between launching a smart contract and a full-blown layer 1 blockchain. We expect many projects to “upgrade” as they grow, and we plan to offer very clean migration paths to our users.

Marketing will also be crucial. While EVM-based rollups have heavily invested in marketing, Cosmos has a reputation for lagging in this area. To justify the development costs of Halo, substantial marketing efforts will be necessary to drive adoption and ensure the investment of community funds is worthwhile. Low-cost rollup transactions will only turn profitable if large transaction volumes are attracted.

We fully support the structured approach proposed for integrating Halo into the Cosmos Hub’s broader product suite. This strategy is smart, and we are aligned with that vision. We simply request more precision on the effective methods that could be proposed.

@Noam

Another key step here would be finding better EVM wallet support for Cosmos chains so that users coming from EVM-based networks are never stuck when they arrive in Cosmos, similar to how Osmosis recently approved Lite EVM. A partnership with the EvmOS team seems like a logical step to take here, and we are already in talks with them about how to offer this to our users.

We absolutely agree with this and recommend the partnership. We encourage the teams to formalize this agreement and include it in the future funding proposal.

---

Summary:

In summary, we fully support this project. The team has proven itself to be highly capable, and we believe they are well-suited to design and implement these proposals. From a strategic perspective, timing and marketing will be key to success. Our analysis suggests that the roadmap should be shortened by at least one quarter to ensure competitive positioning against Ethereum’s SSF evolution. Marketing efforts should also extend beyond the product suite, emphasizing ease of migration into the proposed design.

---

Thank you for reading,
Govmos.

[tknox35]

Very cool. I know similar ideas had come up in the past, but nobody had tried to build it yet.

The real doubt that comes to mind is the possibility of finding PMF, but conceptually, it’s certainly worthy of consideration. One thing I like in particular is that there’s no token involved besides Atom. Anything that’s designed to drive value to Atom gets diluted with a separate token and makes the value proposition much more nebulous.

[Quentin]

More and more composability and modualrity, both in execution and in settlment, I’m not even talking about DA with celestia… also allowing rollups to access IBC is just a game changer, this roadmap seems pretty cool. Here is my understanding of Halo in schema:

In this diagram there are two main routes, the first is the simplest, you are a Halo consumer and you want to both execute your transaction on the Halo sequencer but also settle using the attestors, that’s is obviously the least expensive solution here with a good compromise between smart contracts and ics chains allowing even more modularity for developers on forge.
The other route is more expensive and will surely be developed later,
so here you are an L2 you would like to execute your transactions on Halo to obtain both a reduction in the costs of the sequencers and the possibility of connecting to other networks using IBC and then you settle on Ethereum to obtain the best guarantee. it is important to note that here the whole aspect of the execution is entirely secured by Halo before the settlement, in other words it is a good way to obtain indirect income from IBC for the hub (If you want to connect IBC, you can use Halo as an ETH rollup).
However, we must not forget that Halo will not be the only one to do this and an already existing rollup can at any time use, for example, Malachite (a tendermint like for zk rollups) currently being developed by Informal to decentralize its sequencers, here I am referring to Starknet which should one day be connected to IBC while maintaining its sovereignty and not depending on any other network for its execution.
This is my current take on Halo, thanks for reading!

[BlocksUnited]

Love it! And love the ‘no token’ idea so value accrues to ATOM. The post mentions Celestia. Might Dymension be included? My thought is to benefit as many interchain ecosystems as possible.

[Govmos]

@Quentin

roadmap seems pretty cool. Here is my understanding of Halo in schema:

This is a very insightful representation, and we recommend that the author, @noam include it (or a similar version) in the original post, as it significantly improves the overall understanding of the system and its interacting components.

[Kamikaza731]

Cosmos Hub having it’s own rollup chain is an interesting concept and could possibly work. However I think there is a lot of competition here already that have similar concepts Dymension, Celestia, Initia, Sunrise with some of their own tweaks, adjustments and technologies they use.

These are only chains that are already based on Cosmos SDK there are similar services outside of the Cosmos. So my question here is what Halo does better than the current competition? What will make the developers decide to use Halo over some other service?

To make it clear I do not have anything against Halo it is just with the current timeframe given mainnet might go live in Q3 2025 I just feel like we need some extra edge. I feel like this project will “arrive to late to the party”.

As for the consumer chain how much would it make per rollup? For example how much would an EVM rollup generate revenue per transaction?

[vixcontango]

No new tokens please. More chains but all should be secured by PSS and using the ATOM token.

[Zuko]

This is absolutely FANTASTIC!

I see HALO having a tornado cash like mixer to add privacy feature to Cosmos hub

[seb2point0]

A few random questions:

• Can rollups choose to settle to Ethereum?

• The settlement mechanism proposed sounds similar to Movement’s Fast-Finality Settlement. Does it operate on the same principle?

• What rollup frameworks will Halo support?

• What efforts are Binary Builders or others making to package Halo, Cosmos Hub settlement, Celestia DA, and a rollup framework (i.e.: Rollchains) into an integrated solution for building apps? I think this is important for business development.

[LeonoorsCryptoman]

A long time ago also the model where appchains would route IBC-transactions through the Hub. That would only require that an appchain-Hub channel is available and thereby tokens can be route through the complete Cosmoverse. It would immediately solve the issue of the wrong ibc-denom, since it would always be routed through one and the same IBC-channel. It comes close to the Polkadot-model, with the exception that the appchains can be stand-alone L1’s.
How does that model fit into this proposal?

[serejandmyself]

@Noam

We have submitted a grant proposal to the AADAO to fund this work until the end of year, on which we’re currently waiting for an answer

This wont age well =)

Sorry…

On a serious note, is there a testnet we can join?