# Unlocking New Possibilities with Stateless Architecture in Layer 2 by Leona Hioki | Devcon SEA

- Speakers: [Leona Hioki](https://streameth.org/speakers/leona-hioki)
- Channel: [Devcon](https://streameth.org/devcon)
- Date: 2025-10-09
- Duration: 23:04
- Topics: Science & Technology
- Watch: https://streameth.org/watch/yt-hfj9QaGoZrs
- YouTube: https://www.youtube.com/watch?v=hfj9QaGoZrs

## Description

Explore the potential of stateless architecture in Layer 2 solutions. As Layer 2 technologies evolve, we will discuss the fundamental trade-offs and present how combining client-side Zero-Knowledge Proofs (ZKPs) with stateless architecture enhances efficiency. This session will highlight innovative possibilities not yet widely discussed in the Ethereum community, showing how this approach can revolutionize scalability, security, and privacy.

Speaker(s): Leona Hioki
Skill level: Intermediate
Track: Layer 2
Keywords: statelessness

Follow us: https://twitter.com/efdevcon, https://twitter.com/ethereum, https://warpcast.com/devcon
Learn more about devcon: https://www.devcon.org/
Learn more about ethereum: https://ethereum.org/ 

Visit the https://archive.devcon.org/ to gain access to the entire library of Devcon talks with the ease of filtering, playlists, personalized suggestions, decentralized access on Swarm, IPFS and more.

Devcon is the Ethereum conference for developers, researchers, thinkers, and makers. 
Devcon SEA was held in Bangkok, Thailand on Nov 12 - Nov 15, 2024.
Devcon is organized and presented by the Ethereum Foundation. To find out more, please visit https://ethereum.foundation/

## Transcript

[Music] hi everybody I'm Leona from inmax today I want to talking about statelessness on the how it works on the how how it it behaves and how to cook it so have you ever heard of statelessness or maybe you heard of INX maybe you know know something about her stat lessness I guess so INX is a status layer of etherium and you can believe that or this is surrounding their stateful Network including L one or know the other rollups it's providing their extreme security and privacy at the same time so statelessness is something mysterious sometimes it works like a privacy coin like Monero or gash and sometimes it's like smart contract platform sometimes like a no payment system or app specific chain sometimes General so I want to invite you to the lit Hall of State resonance today um how it works as a transform layer of eum so there's a tradeoff of course now stateless is more and more scalable than stateful lups and stateful earlier ones and it's more private and less functionality so transfer is the most important part of stateless and actually we can do many with this transfer transfer functionalities so if we do the conditional payment we can use a Smart contract to process a payment in certain conditions it makes our decks it makes our nftd Marketplace it makes our yeah many many stuffs are know with that so if we talk about the AR architecture or relationship with the other networks is like no surrounding or know the ethereum ecosystem with stateless payment layer or transfer layer so lups and in ethereums they have more functionality uh with certain State and the conditions and basically INX is more interoperable and it's more Unstoppable um private extremely scalable there why stateless layer is so scalable we can summarize that is because we the by Def it definition it's all about posting the minimized constant data on in a narrow definition of uh in cryptography space or cryptography Financial cryptography the statelessness is all about the minimum data constant minimum data on chain and it meaning that you don't have to annoy by get annoyed by their fluctuation gas because no matter how many people come to in one network it's constant uh no cost on chain right in terms of data cost so surrounding status layer helps their no many use cases no privacy inter properity and constant gas cost eliminating their fluctuate think as we so this is a no clear comparison uh with each other so we can describe this our tradeoff between uh statefulness and the statelessness in a more more detailed way so there are many scalability solutions right so for example in Max 2 and plasma NEX is is coming from inmax it's more focusing on statelessness Lightning network is the most or famous or scalability solution in Bitcoin space um plasma is well know discussed are in are five years five years ago or yeah something like that 2018s and lups it's buring and the maturing right now right so in terms of this comparison are I don't say it's kind of 100% Fair because sometimes know we need to other privacy which lighting Network and INX has have sometimes our lups have more functionality so we need add that so yeah I don't think this our triangle describes everything but basically in terms of most important part know the basic us basic safety and the scalability it's all about statelessness and the capital efficiency and offline safety if you have a offline safety you can have no you don't have to run node to keep yourself safe and if you have Capital efficiency you can yeah no spend your all money instantly after you deposit right and statelessness always it's all about scity the posting their minimize our data chain so this triangle is quite important to have the essential uh scalability of blockchain and in terms of that yeah inmax 2 plasma X have a very good triangle and the lightning network uh it has a it's a great scalability but uh Capital efficiency and offline safety is quite no challenging and rollups they have a quite good ux with capital efficiency and good offline safety but it's less scalable oops oh okay so yeah I want to introduce briefly uh how to cook and how to cook statelessness it's it's like first step one uh we need to have a communication between block producers on the user site user site receives their Mar proof other utxo so e each micro proof given by operator or block producer is utxo and utxo itself is a yeah micro proof and you can input that in their no Z circuit offchain and the basically you can uh sign back to the blocker producer to yeah prevent uh skipping this uh procedure then basically you it Mak making sure that uh you guys can have your data uh on your client side then yeah this is how to make her know safety in this communication compare with zkp and our block producing onchain yeah so proof update no it's like no uh in this communication and once you you are given proof given the proof you need to prove that your proof is not spent after that right and basically we can apply the no transaction uh proof after you no you were given there no micro proof or no by block producer then you can prove that you didn't spend your money after no you you got this proof then this proof is still valid after your block is after the block is changed or after we have a new block or state so surprisingly basically it makes a only five bytes data on chain and for each transaction and the each transaction you can increase a 10,000 uh transfer uh to to that five by transaction cost without any additional data now this is a quite a big achievement we had for for this to make this stateless payment layer and how to make sure that uh your utx or given mro proof is safe no you can prove that no it's not spent after that but your kind of incoming transaction or incoming utx so how how to make sure that you you know it's safe and it's true money it's a interest chain of the balance proof right no your utxo is coming from another person and another person got that from another person right it's the endless chain of transition uh of UD XO or the ownership of UD EXO right so we can so in some client ver verification or client side client side or validation project however no infinite lengths of this verification actually and how to make it efficient enough to be practical so we can obviously apply the recursive zkp right this is the endless recursive zkp so in this architecture so basically you can see thatp n minus 2 is Prov n minus one and the DPN minus one is proven in our DPN and and this is a end chain of zkp right so once once you verify the previous zkp no this proves the entire history of DGP at the same time right so because no yeah it's already proven the previous one and the previous one already proven the previous one previous previous one right so this chain of zgp is extremely or extremely efficient way to verify the entire history of XO once you verify the previous one no simultaneously it mean that no you verify and prove that entire history of UD XO and the entire history of zgb chain is valid so coming back to the previous slide it's the endless chain of zkp and the endless chain of the utxo passed uh offchain nothing go on chain and the this goes simultaneously on each client side so if there are million users million label priorization if there are billion users billion label priorization happen in each client s so yeah I want to make it short so yeah the last advertisement is like yeah one transaction you can increase an no almost unlimited number of transfers uh in one transaction without adding uh any uh onine data basically it's constant and you can Pro yeah no aggregate so many transactions and everything will be totally private and everything will be totally peer-to-peer so this is a no construction and the surprising uh result which we made and at the last we need say that how we base roll up and what is base rollup yeah basically Justin Drake come up with this idea and according to him uh we our team found the concept of Base rollup independently a bit later than him and basically INX to is base rollup and base rollup means that it's interoperable with the other rollups uh especially the other D rollups uh base rollups so basically concept of Base rollup is is making many many layer tools one big layer to and it's the most yeah recently I found that is the most exciting idea in this space as well and yeah base Roll Up is like uh yeah uniting their mechanism uh in layer to space and INX is originally uh inmax 2 is originally uh based rollup meaning that it's interoperable with the other chain and the once you verify uh or send their transaction uh in this INX 2 rollup you can transact with the other smart contract on the other uh base roll up and yeah it's more like a academic side or no take on the statelessness I I said so many things about the statelessness and the I said it's great and at the same time there's a criticism that are it's impossible and basically they what they are saying is totally correct and they are saying that uh this lead lines and basically it's meaning that let's say stateless system needs frequent local proof update and the local proof update uh is some sometimes it's quite annoying and sometimes it's not so annoying or it's quite harmless and in our architecture local proof update is quite quite uh harmless so yeah basically the main direction to solve this problem is like making local Pro updates less harmful or CR critical or totally harmless and in our system it's almost uh yeah it's just make a five second delay uh in in our communication uh between block producer and the transaction cender so basically it's are quite solved in our no uh development but yeah please remember that uh stateless is kind of possible uh in some conditions and it's inmax is avoiding this program okay at the last uh we are doing the Privacy mining right now so in I said like uh stateless architecture have a strong feature of privacy and it brings the Privacy to layer one at the same time right now so once you bettery uh once you deposit money to our contract and withdraw our contract basically we we don't have any links between them nobody can see the links between them right and we l in that are no mechanism and the more and more people come into this privacy mining uh more and more anonymity set the greater anonymity that we can have are know together right so it's a very big uh social experimen for me to do uh simply because it's something new to kind of ening people they're know making people able to buy anonymity set right and it's a chance to for the market to buy anonymity set or instead of or no paying their no transaction fee or in their no privacy uh protocols and it's a new kind of way to uh find the P find the find the price of privacy and we cannot Define their f price of privacy today but it's a experiment to make a equilibrium in their Market to to find out how much do we need to pay for their anonymity set so please join this social experiment on the of course get Li world that and the yeah some please make your asset private in this program so this is a just a invitation on this a bit independent from their no stateless or stuff but yeah I'm doing this or passionately so please join this and thanks for listening to me today and yeah sir [Applause] down all right thank you so much Leona uh we have one question so far so please don't forget if you have any questions for Leona please uh scan the QR code and answer the question but let's start with this one so given the huge difference with General l2s is the expectation like users to use others usually for executing contracts and spending inmax when they want to Simply transfer a lot and or with privacy so yeah I a bit are know are taking time to understand this question but basically uh yeah of course it's a huge difference between the general or no contract execution in other layer tools and the know basically sending the token inmax and the inmax are provider conditional payment so sometimes it's make our our no execution of smart contract but it's depend kind of no it's restricted to uh conditional payments which means that you cannot lock the token uh on Smart contract like AVI or no Unis swap on these kind of stuff and it's more of course it's private and so expectation so so basically we are expecting that uh no people use this for extremely uh cheap payments with uh smart contract condition and yeah that's their expectation so yeah I think it's quite our correct our understanding okay thank you and the second question would be how to manage higher computation requirements and potential latency in transactions so higher computation requirements and yeah this is quite a good question and the actually is quite limited um so as I said as I described on the chain of the zp or no picture basically each zkp is are proving the entire history and it's constant cost and it takes are 2 seconds in Mark M2 and the 5 seconds are in older version I guess um it's quite limited so yeah higher computation requirements are it's it does it's not true actually and yeah basically we can send that or two or five seconds or delay or latency is there everything or we can have other no we we need to have other downside and yeah it's quite limited and the at the same time it's getting shorter and shorter because of the no evolution of machine on the zkp itself so yeah the latency five second latency makes everything scity inity privacy and this kind of stuff okay someone was able to sneak in another question so we have how privacy mining connected to the statelessness and inmex I would probably say how is privacy mining connected to the statelessness and inmax so it's all about are making great anonymity set and so statelessness are actually it's not so connected with statelessness statelessness makes privacy and make their breach other privacy application right so for L1 users no inmax works like a no privacy application and the for the payment user or L do users it's uh it's like a no payment Focus up specific L do right so I think that uh yeah the only the connection is there no privacy and we need a more greater set of anonymity in terms of privacy so that's a reason why it's somehow going connected and we need to yeah emphasize that this token economics uh is quite useful uh for both of them L1 users and another L2 users all right now you come with the questions we can only have one last one and that is and only a quick answer how to bridge esc20 token to inmax oh just deposit it's no that's it doesn't work all right then let's wrap it up give you a big Applause to Leona [Applause]