Ethereum Core Developer @nethermind Opinions are my own. believe in somETHing.
kuwait
Joined December 2009
- Tweets5.7K
- Following458
- Followers1.1K
- Likes7.6K
smartprogrammer.gwei 🐌 retweeted
Over the past few months, the @Nethermind Client team has been focusing on almost everything BUT performance - authoring, supporting and reviewing EIPs for Glamsterdam, championing Hegota headliner - Frame Transactions, researching state bloat, and much more.
That's not changing - we are still being heavily involved in Ethereum roadmap. But AI has unlocked a whole new level of productivity for us, and we're now pushing hard on several fronts:
1. zk guest program - coming soon to @eth_proofs. Very promising, very performant.
2. RPC refresh - Geth compatibility, major performance improvements, and more. Few projects already got a feeling of what we are shipping - and it is looking very good.
3. Client hardening with AI - Node MCP for users and our custom CoreMCP for developers, helping us ship faster and build a much more stable client.
4. Innovative node optimizations and advanced plugins - unlocking capabilities that could be extremely useful for many projects across the ecosystem, including support for new chains.
And that's just the beginning.
As a "side effect," we've also significantly improved the node's performance at the tip of the chain - before even getting into the Glamsterdam hard fork, which should unlock a whole new level of throughput.
Nethermind is the fastest #Ethereum execution client, and the only one pushing beyond 1.5 GGas/s in p50 block throughput at the tip of the chain.
With the next fork, we expect that ceiling to go even higher - based on Sepolia data, potentially around 3x higher.
And we're definitely not done.
We want to change the rules of the game. The AI-powered tools we've built in-house are giving us enormous leverage, and we're confident this advantage will only grow over time.
Performance is good. 🚀
GLM-5.3-Flash-EXL3 vs DeepSeek-v4.1-Flash-1017-EXl3 on 2 DGX Spark, TP=2. Seems benchmarks are truly strong. time to test it on real work. Thanks @MiaAI_lab for the recipe and SparkDash.
smartprogrammer.gwei 🐌 retweeted
We are considering removing our precompiles dependencies and reimplement them in C# as we can do it faster, avoid managed<->unmanaged code cost, avoid 3rd party dependencies and increase diversity (not all clients depending on same library that can have security issue).
smartprogrammer.gwei 🐌 retweeted
I built it instead 😁
Introducing LeanCIL - formal semantics for .NET CIL in Lean 4, with machine-checked proofs against actual compiled assemblies
Not just C# - a foundation for verifying F# and other .NET languages too
Let's formally verify dotnet
github.com/NethermindEth/Lea…
I just proposed an official @leanprover 4 formal semantics for C# - giving C# a foundation for machine-checkable proofs.
AI agents are getting very good at writing code - and very good at breaking it.
In the agentic era, "tests passed" won't be enough for critical software. We need to be able to prove properties of the code we ship.
And this is becoming practical.
Rust already has Aeneas -> Lean. Microsoft Research is using it to formally verify SymCrypt. Java and C have roots into formal verification too.
C# should not be left behind.
AI can write the code.
AI can construct the proof.
A deterministic proof kernel checks whether the proof is actually correct.
That is a very different trust model from:
AI writes code -> AI reviews code -> hopefully both were right.
Proposal:
github.com/dotnet/csharplang…
smartprogrammer.gwei 🐌 retweeted
I have a games industry background, where everything needs to fit within 16.67ms to hit 60fps
That mindset has carried over into other work
So now @Nethermind processes Ethereum mainnet blocks in 16ms at p50, on standard EIP-7870 hardware
smartprogrammer.gwei 🐌 retweeted
"lead and adopt the same changes" in practice.
github.com/paradigmxyz/reth/…
Less than 12h between @Nethermind PR merged and Reth PR opened.
Hard to be unique in AI times - we expected that :)
What matters is who brings the innovation and that should be higlighted.
This might not be widely known, but @Nethermind improvements don’t just benefit our client.
Other clients often follow our lead and adopt the same changes, so we’re raising the performance bar across the entire Ethereum and EVM ecosystem. 👇
smartprogrammer.gwei 🐌 retweeted
This might not be widely known, but @Nethermind improvements don’t just benefit our client.
Other clients often follow our lead and adopt the same changes, so we’re raising the performance bar across the entire Ethereum and EVM ecosystem. 👇
Nethermind continues doing excellent work and making Ethereum more efficient and faster.
firefly.social/post/x/210738…
smartprogrammer.gwei 🐌 retweeted
The fastest.
we just made @ethrex_client again the fastest client in @ethereum. i've always said, lambda would deliver and push the boundaries of ethereum.
after a couple of months focused on Glamsterdam hardening and Hegota prototyping, including hosting several public testnets for the most important EIPs coming in Hegota, we've turned our attention back to performance. this doesn't mean we're stepping away from that work, where we still have a lot to contribute.
The early results are very promising. In just a couple of days, we brought block processing time down from ~40ms to ~26ms in mainnet, making ethrex the fastest client once again. The image below compares how we perform today (ethrex-baseline) with what we've been cooking (ethrex-testing).
congrats @class_lambda, i'm proud of the team we've built. i love working with people that are way smarter than me.
thanks @ethereumfndn for the constant support.
smartprogrammer.gwei 🐌 retweeted
BAL's + @Nethermind crazy optimization in action on Sepolia. Expect even bigger throughput numbers when Block sizes rise up to 200MGas (and more!)
smartprogrammer.gwei 🐌 retweeted
Nethermind continues doing excellent work and making Ethereum more efficient and faster.
firefly.social/post/x/210738…
smartprogrammer.gwei 🐌 retweeted
Nethermind Beacon Chain plugin ready and synced for the Sepolia fork 🚀
Nice bonus: running EL + CL as one app doesn't seem to add much memory overhead
smartprogrammer.gwei 🐌 retweeted
Some heavy load on mainnet in last 6h, but our node didn't even budge
smartprogrammer.gwei 🐌 retweeted
Mandatory upgrade for Sepolia nodes: run Nethermind Client 2.1.0 before the Glamsterdam fork on Oct 6. Your CL needs it too.
On flat archives, 2.1 adds an opt-in index that cuts single-transaction traces from ~7.1 s to ~28 ms, ~250× faster in our tests.
github.com/NethermindEth/net…
smartprogrammer.gwei 🐌 retweeted
Apparently this is circle jerking? ethereumuptime.com
ethereum became weak because many core devs are just circle jerking themselves. its strange for them to hear a different voice. they became so used to being all friends that grt payed by the same people that they see criticism as strange.
having opinions that i believe are the best for the network means for some of them that i am trying to capture ethereum. before this was called the social layer. monoculture maxxxing.
if you want to win an argument just counterargument. saying i’m just a player with no morals that is selfish is useless in a debate
smartprogrammer.gwei 🐌 retweeted
C# FTW again github.com/devmobasa/ttfx
Against Python up to 140x faster!
Against Rust from 0.3.3 version that DHH referenced here up to 2.12x
Since you don't like Rust, you can just port it to a much prettier language like C#, and no need to sacrifice performance.
C# on Linux has been such a bliss. So good, and so fast.
For bash-like scripts, tools, CLIs, services, apps.
I ported the Omarchy screensaver engine (ttfx) from Rust to x86-64 assembler, and it's up to 17x faster!! One-shot translation by Opus 5.5. We keep drilling until the agentic drill bit hits bedrock! github.com/omacom/ttfx/pull/…
smartprogrammer.gwei 🐌 retweeted
Nethermind Client 2.0.0 is out. It changes how the node stores Ethereum's state: archive nodes drop from ~30 TB to ~2 TB, you choose the history you keep, and blocks process 2–3× faster than the previous release on a fresh node.
Nine years since the first commit. The fastest execution software we've ever built.
Breaking changes; read the notes before upgrading. Release: github.com/NethermindEth/net…
Btw, @Brechtpd i run local model on my home server, i wouldnt mind connecting the app to it instead of running on the phone and burning battery. So maybe consider an openAi compatible url with an api key in the settings for people that want that 🙏