RT @CPerezz19: BloatNet has already been setup! And we aim to start running the first round of sync tests this week!. Make sure to stay in….

RT @lightclients: As of today, all Ethereum execution clients support history pruning for pre-merge data. For mainnet, this means 300-500 G….

16/ For the full deep dive, watch the talk by @gballet in ETHCC:. 🎥

15/ The future is likely *partially* stateless first—via VOPS, substate clients, and code chunking. Tree change + ZK proving will take time—but the foundation is being laid today.

14/ BloatNet helps identify breakpoints in sync, execution, and storage. It’s a testing ground for upcoming EIPs and allows a data-driven approach to protocol evolution.

13/ So when will statelessness actually ship?. We’ve launched BloatNet—an initiative led by @CPerezz19 and @ethPandaOps partnered with @NethermindEth to stress-test Ethereum by intentionally bloating the state. 👉 Check out

12/ But current bottlenecks may lie deeper—in how the state tree is encoded in the database. Hence, EIP7736 isn't the perfect solution either. Nonetheless, this area is actively being researched—led by @ngweihan_eth.

11/ State Expiry targets the long tail of Ethereum state—most of which is stale. EIP-7736 proposes pruning old, unused state in Verkle Tree to reduce storage burden and increase throughput.

10/ Why? Real-time proving today demands costly machines—and there’s no strong incentive for independent provers (yet).

9/ ZK Stateless is often confused with Verkle. No, ZK Stateless is not Verkle trees!. Owing to the huge computing power requirements, ZK Stateless designs are currently fully out of protocol, and introduce a centralization risk.

8/ Arbitrary substate: not all nodes need the full state. Operators (e.g. dApps) with predictable access patterns can selectively retain only relevant state—such as contracts their app interacts with. This reduces storage without compromising function.

7/ VOPS (Validity-Only Partial Statelessness): clients retain *minimal* state. VOPS clients discard trie data and storage slots—keeping just accounts. This enables lightweight nodes with specified protocol roles, e.g. building inclusion lists, participation in rainbow staking.

6/ Between those extremes lie several hybrid solutions:. - VOPS.- Arbitrary substate.- ZK Stateless.- State Expiry. Let’s break them down 👇.

5/ When evaluating stateless solutions, we need to consider three axes of design tradeoffs:. - In-protocol ↔ Out-of-protocol.- Stateless ↔ Stateful.- Decentralized block building ↔ Centralized block building.

4/ With code chunking (EIP-2926), we can break bytecode into smaller chunks, load them on-demand, and only prove the chunks that were accessed. This keeps memory and proof sizes bounded—even with large contracts.

3/ While we explore binary trees long-term, one feature from Verkle can benefit the protocol *right now*: code chunking. Today’s code size limits are too small for real-world developers. But lifting them causes problems for ZK—more RAM, disk, and massive proving overhead.

2/ For the past few years, Verkle trees have been the answer for stateless clients: smaller trees, smaller inclusion proofs, and ZK-friendly (for EC-based proving systems). But binary trees are making a strong comeback—they work better with STARKs and offer quantum resistance.

🧵1/ Statelessness is here to stay—and it’s evolving faster than ever. Here’s what’s changed, why it matters, and where we’re headed 👇.

RT @jbaylina: At @ziskvm, we’re now proving Ethereum blocks in real time — open source and running on increasingly efficient infrastructure….

RT @asanso: Happy to share joint work with Giuseppe Vitto we presented a few weeks ago in Madrid (soon on ePrint)! We leverage the Graeffe….