Velarium

Making blockchain execution faster, smarter, and ready for what's next. Early stage.

[Velarium @VelariumOne]

We’ve built the fastest RISC-V engine we could find. So we made the benchmark public. Don’t believe us? Run it: https://t.co/QBk3lQiJGG

[Velarium @VelariumOne]

Thrilled to announce @Be1garat as an official advisor to @VelariumOne.

Former SVP at @MatterLabs, Director at the ZKsync Foundation, and the person behind one of the largest raises in the Ethereum ecosystem.

Big things incoming. More soon. 🔧

[Velarium @VelariumOne]

Big news for us! We're excited to welcome @anthonykrose as an official advisor to Velarium.

Anthony is one of the most respected engineering leaders in the Ethereum ecosystem and brings a deep network in this space.

Having him on board is a huge step forward for us as we build!

[Velarium @VelariumOne]

We've got our first results for running an EVM erc20 contracts.

7x faster than revm.

Not all optimizations are in, we're confident at 10x, possibly faster.

Once the compiler is hardened enough, we'll release a public benchmark harness to prove this.

[Velarium @VelariumOne]

@Defimarvel01 Hey @Defimarvel01, thanks for your comment, it's great to see!

We'd be interested to chat more - please could you email us at [email protected]?

We can arrange a time to meet.

[Velarium @VelariumOne]

Today we’re revealing Velarium.

A parallel execution layer for Ethereum chains - built to make existing stacks faster, without changing how they work.

Here's what we're building and why it matters 🧵👇

[Velarium @VelariumOne]

3/ Microbenchmarks are easy to abuse.

Still, they’re useful for one thing:
checking whether the architecture cashes the check it wrote.

This one does.

[Velarium @VelariumOne]

1/ Got our first microbenchmark result: end-to-end execution of a tiny contract
(CALLDATASIZE + 2× CALLDATALOAD + ADD + MSTORE + RETURN)
in ~12 ns.

Faster than a RAM access.

But “faster contract” is only half the story.

[Velarium @VelariumOne]

2/ An execution overhead this low unlocks scheduling techniques that are simply not viable otherwise.

[Velarium @VelariumOne]

With the @ethereum Strawmap, L1 targets 10k TPS (Gigagas) by ’29.

When it lands, a lot of L2s lose the “we’re faster” wedge unless they can sustain 5–10× that throughput.

Who’s ready for 50k+ TPS execution→witness pipelines?

[Velarium @VelariumOne]

@OluwafemiFate @velarium Thanks @OluwafemiFate

We plan to have a public benchmark harness on a couple of months. Till then we'll provide teasers and will be adding info to the website. If you're interested in something in particular - ask away.

[Velarium @VelariumOne]

We're early stage - verifying architecture, building traction, and looking to work with L2s and EVM-compatible chains who want to push throughput limits.

If that's you, or you know someone it should be - share this thread.

DMs open. 👇 https://t.co/JE0Ldklmqb

[Velarium @VelariumOne]

Two ex-Matter Labs alumni decided there was a better way to push EVM throughput

@AiKiXD and @GeCumbo - we've spent the last few months heads-down validating the architecture and building the proof of concept.

No token. No new chain. Just execution, done better.