In case anyone missed it, yesterday started the EF devnet 5 client interop runs. We already merged our devnet5 branch to main and are running internal devnets in parallel, looking for issues or performance improvement opportunities.

running SHADOW SIMULATION of zeam devnet5 multi node network 100% 28 core loading 200% 96GB mem DDR5 7200MT/s loading Disk IO engaged on nvme gen5 samsung disk beautiful

📢📢📢 we are now devnet5 interop!!! 🚀🚀🚀 which is the culmination of the PQ consensus stack that we have been building and iterating on for @ethereum mainnet. PQ signatures are different from BLS signatures, the current signature machinery of ethereum. But BLS signatures are not Post Quantum secure because breaking elliptic curve cryptography is not an exponential problem for Quantum Computers. But ethereum is being build to last centuries not just decades. To that end @leanEthereum has been working on PQ signature cryptography using hash based signatures which are one time signatures (OTS), and not plainly aggregatable. Not only that they are huge (~1.5kb). So this entire challenge started the "devnets" initiatives of leanEthereum. Herein come's the "leanVM" the ZK rail which can aggregate such signatures and makes the entire PQ strategy possible. We have already been through devnet0 to devnet4, and now devnet5!!! Devnet5 is monumental in that regard, entire block will carry just 1 signature, all aggregated across packed attestations, block signatures (and anything else that will comeup when we backport the spec to mainnet ethereum) However this is one side of the puzzle, to maintain a stable node, one should be able to repack the attestations from a side branch especially if it moves justification and finalization. and Voila again with leanVM magic, we are able to split the attestations that we need to repack from the combined block signature and repack/re-aggregate them into a new block proposer wants to propose. This places leanVM as the centrepiece in the entire ethereum post quantum strategy. and the current aim of all the devnets we have been running is to bring a production level performance and demonstration of the capabilities. There are 8 clients that participate in these devnets each bringing value to the table to add stability and robustness (and chaos lol). Because we know: "There's many a slip between the cup and the lip" Goal of all leanEthereum clients is to remove them, one "slip" at a time (or multiple slips at a time lol). Spec isn't good enough, we need production performance, Spec and production design/performance are unequivocally tied. this isn't just a POC network, this is a proposal for ethereum mainnet! Thats why we have been rigorously running devents, slowly scaling the validators and subnets and discovering and alleviating the issues so that we end up with a production grade PQ signature scheme that ethereum deserves and needs. And this focus is now gonna magnify 1000X now that we believe we are on a spec that can deliver PQ for mainnet. And mind you, the time of need is gonna strike soon. We not only intend to solve this conundrum for ethereum but propose to even upstream to bitcoin so that we have a PQ standard this entire space deserves. And all that based on a humble but extremely powerful leanVM that makes the hashbased cryptography workable for the production grade systems like ethereum and bitcoin. PS: we are currently heavily focused on debugging and scaling devnet4 spec devnets while we have already started to run sims and preliminary interops for devnet5. So stay tuned for further progress that we leanEthereum teams have been cranking out with a steady but heavy dose ☕️. May be all matrix is just ☕️☕️☕️

ETH Daily - 29th May 📰 -Vyper vyupgrade tool 🛠️ -Lean Devnet5 interop ✅ -EthPandaOps simulates FCR 🧪 -Wonderland unveils DARC 🔎 Presented by @ethconf 🗽 ethdaily.io/957

One day, quantum computers will be able to crack the cryptography that secures almost every blockchain. Most of crypto is ignoring that. Ethereum isn't, where devs are constantly cooking on the fix. @leanEthereum's devnet5 is a big step. A whole block's worth of the new quantum-proof signatures, now squeezed into one.

🎉🎉🎉 devnet5 spec merged 🚀🚀🚀: block with just a single aggregated signature proof github.com/leanEthereum/lean… this brings @leanEthereum full circle on the "PQ stack" for consensus to upstream it to mainnet PQ centric L* hardfork It includes a very neat feature for splitting the block signature proof for re-bundling it on a side branch even without requiring the constituent signatures that were aggregated! all thanks to awesome EF researchers working on leanVM 🦾🦾🦾 now time to build devnet5! comeon lean teams time to harden and show production stability and scale fyi: devnet4 runs are already on for scaling the validators to harden subnets & aggregators lets grind on!

since we are closer to devnet4 spec freeze and interop 🚀🚀🚀 here is our devnet4 summary and devnet5 proposal for @leanEthereum which we will put forward discuss in today's PQ Interop call 👇👇👇 hackmd.io/@qYrlZEprQ1iz7Njii… all client teams pls take notice 😃🙏

#Ethereum #Fusaka yükseltmesi #Pectra gibi resmi olarak ertelendi! Ana ağ aktivasyonu, Kasım yerine 3 Aralık 2025'te gerçekleşeceğini söylüyorlar fakat başka sorunlar çıkacağı şimdiden belli oldu. Bu karar, "All Core Developers Consensus" (ACDC) 165. toplantısında alındı. Testnet'lerde ki hatalar ve veri kapasitesi ölçümlerinin yetersizliği. Ethereum'un Pectra sonrası büyük hard fork'u. Ana odak: Veri kullanılabilirliği (DA) iyileştirmeleriyle Layer-2 (L2) ölçeklenebilirliğini artırmak. İçerdiği 11 EIP (Ethereum Improvement Proposal) ile ağın kapasitesini genişletecek. Ana yenilik: PeerDAS (Peer Data Availability Sampling). #PeerDAS detayı: Blob'ların (veri paketleri) ağda daha verimli dağılımı. #Node'lar tüm blob'ları indirmek zorunda kalmayacak; sampling ile veri bütünlüğünü doğrulayacak. Bu, full node'ların yükünü azaltırken, sansür direncini artıracak. Erteleme nedenleri ise şu: Fusaka #Devnet5'te #blob kapasitesi testleri başarısız olduğunu gördüm. Geliştiriciler (#Nethermind, #Besu, #Erigon vb.) eth'den ek süre istedi. Ekim'de üç testnet (#Holesky, #Sepolia ve #Ephemery) üzerinde kamu testleri başlayacak ve 2M$ denetim programı devreye alınarak testler hızlandırılacak. Bu iyi haber. Blob kapasitesi artışı: - Fusaka ile hemen olmasa da Upgrade sonrası iki hard fork ile kademeli olarak İlk BPO (Blob Parameter Only) fork (17 Aralık'ta) blob/block'ı 10/15'e çıkaracak. - İkinci (Ocak 2026'da) 14/21'e. Maksimum 128 blob'a kadar yol haritasında bir fikri birliği var ve çıkarılması planlanıyor. Diğer EIP'ler: EIP-7762 (EOF v2), EIP-7623 (gas ücretleri optimizasyonu ki bize lazım olan bu kısım!), EIP-7639 (blob ücretleri ayarı) vb. Bunlar, calldata maliyetlerini düşürerek rollup'ları (#Optimism, #Arbitrum) daha verimli kılacak. Etkilerleri: - Pozitif - L2 ücretleri P düşebilir, ağ daha stabil. - Negatif - Kısa vadeli piyasa dalgalanması, geliştiriciler için timeline kayması. - Teknik derinlik için: PeerDAS, Danksharding'e giden yolun parçası olarak görülüyor ve Eth3.0 için yani Beam/Lean Chain için gereklilik olarak görülüyor yani bu güncelleme olmazsa olmaz. Danksharding ile de 30MB veri/bloklama mümkün olacak. Bu sayede Bloklama ve Bloblama yükselecek ve ağ ve gas/fee optimizasyonu sağlanacak. Fusaka, proto-danksharding'i (EIP-4844) genişletilmesi en önemli unsurlardan bir tanesi. Not: Pectra da 2024'ten 2025'e tam 1 yıl ertelenmişti. Bana göre Fusaka güncellemesi de aynı kaderi paylaşacak gibi duruyor. Hal böyle olunca Beam/Lean Chain geçişi 2028 olarak planlanmıştı fakat bu aksaklıklar neticesinde 2029'a hatta 2030'a kadar sarkabilir. Daha #glamsterdam güncellemesi var 2026 da buda 2027 sekerse cümbüş başlar. Korkuyorum ki uzun süre uykuda çile çektirebilir. Umarım tersi dış hareketler gelir yatırımcı yüzü güler. Hadi geçmiş olsun. .... finance.yahoo.com/news/ether… cointribune.com/en/ethereum-… coincentral.com/ethereum-fus… cointelegraph.com/news/ether… coindesk.com/business/2025/0…

🛠️ DevLog – DevNet Restructure & Testnet-0 Setup We've officially moved forward with setting up Testnet-0 on Arbitrum Sepolia (L2), and are phasing out the previously considered DevNet7. This marks a shift to a more streamlined and production-aligned flow. 🔹 Updated Network Structure: - Devnets (Feature Testing – Internal): devnet-0 – Active (likely reuse of DevNet5) devnet-1 – Active (DevNet6) - Testnets (Staging / Pre-Mainnet): testnet-0: Arbitrum Sepolia (L2) ✅ Now Live 👉 dashboard-testnet0.cortensor… testnet-1: L3 COR Rollup (COR gas) 👉 dashboard-testnet1.cortensor… - Mainnets (Upcoming): mainnet-lite: Arbitrum L2 mainnet: L3 COR Rollup (gas-native) 🔁 New Dev-to-Mainnet Flow: Feature → DevNet → Testnet-0 / Testnet-1 → Mainnet (Lite or L3) This structure simplifies iteration and staging, giving developers and node operators a clear view of where each module lives and how it progresses toward production. With Testnet-0 now bootstrapped, we'll focus next on finalizing remaining components and staging towards mainnet. #Cortensor #AI #DePIN #DevLog #Testnet

🛠️ DevLog – DevNet6 Migration Reminder & Critical Fix 🔹 Please Migrate to DevNet6 DevNet6 is now live - all nodes should upgrade. Migration is simple: update your runtime and restart. Latest switch script is available in the installer repo: github.com/cortensor/install… 🔹 Critical Bug Fixed A major bug affecting precommit retry logic on network tasks has been fixed. This fix is included in the latest binary and installer version. DevNet5 nodes will not receive this fix - migration is required for proper task execution. Keep your node updated and help us stabilize the network for testnet readiness. #Cortensor #DevNet6 #Migration #DevLog #AIInfra #DePIN

📸 Phase #4 – Snapshot & Prize Eligibility Update We're finalizing the snapshot and processing the eligible node list. The official prize & role list will be released within 24 hours. 🔹 Final Top 150 will be locked for reward distribution 🔹 Do not shut down your node - offline = disqualified 🔹 Continuous uptime is required for final eligibility 🔹 DevNet6 migration instructions coming soon Thanks to all node ops for pushing Phase #4 to new heights. #Cortensor #DevNet5 #Phase4 #NodeOps #DePIN #AIInfra #TestnetPrep

🧪 Phase #4 Recap – Trustless AI Infra Under Load Phase #4 was Cortensor's most advanced testing phase yet - pushing live task routing, payment flows, multi-oracle sync, and node diversity under real conditions. 🔹 Live User Task Pipeline End-to-end tested: session deposit → credit allocation → task execution → node payout via $COR. First user task payments confirmed on-chain. 🔹 Internal Mini-Hackathon Five working apps built in two weeks, surfacing critical edge cases and driving UX fixes across the task pipeline. 🔹 User Task Flow Polishing Internal testing helped refine session states (ack → precommit → commit), ranking logic, and failure handling. 🔹 Dynamic Node Shuffling Randomized miner selection through precommit rotation improved fault tolerance and load distribution. 🔹 Router Node REST API Reverse Proxy Apps tested against production-ready API with full SSL support, key-based auth, raw prompts, and flexible proxy routing. 🔹 Multi-Oracle Coordination 2–4 oracles per network with 15-minute rotation cycles now live for sharded validation and leader election. 🔹 SessionPayment Module Tested in a loose OracleV3-triggered mode for credit and payout flows. Will evolve toward tighter contract-to-contract logic. 🔹 CognitiveLevel PoC Deployed on DevNet6. Begins groundwork for SLA, model-aware routing, and node classification by performance. 🔹 NodeVersion Tracking New module added to track historical node versions for better debugging, filtering, and upgrade management. 🔹 DevNet6 Migration All major modules moved to DevNet6 for deeper iteration and Phase #5 preparation. 📊 Stats: dashboard-devnet5.cortensor.… Phase #5 will tighten, automate, and scale toward a resilient testnet. #Cortensor #DevNet5 #Phase4 #AIInfra #DePIN #MiniHackathon #UserTasks #E2ETestin

🗓️ Weekly Focus – DevNet5 Snapshot, DevNet6 Iterations, and Hackathon Highlights 🔹 Phase #4 snapshot completed – final prize list and reward calculations are underway this week. We're reviewing session/task data to ensure accuracy and fairness. 🔹 Our 1st Internal Mini-Hackathon wrapped alongside Phase #4 – five functional community-built apps were submitted. This week, we'll spotlight each project through dedicated social posts. 🔹 CognitiveLevel remains in the PoC and design phase – early integration and testing underway to support node classification, SLA enforcement, and adaptive task routing. 🔹 OracleV3 and MinerV4 enhancements underway – adding more robust error handling with try-catch and retry logic for improved reliability under edge cases. 🔹 Another iteration on SessionAuth is in progress – addressing gaps from the previous cycle to ensure stronger authentication between modules. 🔹 Dedicated GPU nodes will be configured to keep top showcase apps live – ensuring consistent performance as we continue to iterate on SLA and CognitiveLevel features. 🔹 SessionPayment Iteration #2 – further testing payout triggers via OracleV3 events. This staged approach lets us validate the flow before enforcing it contract-to-contract. 🔹 Phase #5 planning is active on DevNet6 – focusing on reinforcing payment logic, coordination layers, and node reliability, moving us closer to testnet readiness. Let's keep building toward a robust decentralized AI network. #Cortensor #WeeklyFocus #DevNet6 #Phase5

Reminder: DevNet #5 Snapshot Incoming The final snapshot for DevNet #5 will be taken within 24 hours. Make sure your nodes are active and properly configured - this is your last chance to secure your position before rewards are calculated. ⚠️ Important: Stopping your node immediately after the snapshot may lead to disqualification. Keep it running. #Cortensor #DevNet5 #NodeOps #DePIN #AIInfrastructure