Solana Firedancer client now live on testnet

Solana’s second unique validator client, Firedancer, is widely said to be more performant than the Agave client

article-image

Akif CUBUK/Shutterstock modified by Blockworks

share


Today, enjoy the Lightspeed newsletter on Blockworks.co. Tomorrow, get the news delivered directly to your inbox. Subscribe to the Lightspeed newsletter


At long last, the much-hyped Solana Firedancer client appears to be boogieing its way into the light. 

Jump Trading chief scientist Kevin Bowers announced it in about the nerdiest way possible at Solana Breakpoint, pointing out to a crowd of onlookers that a string of numbers indicating a software version on Solana mainnet looked suspiciously unlike Solana’s current clients. This led to some applause. 

Firedancer is an independent validator client, or software that validators can run to build the blockchain, being developed by Jump. The client is being written independently of the original Solana Labs validator client, now called Agave (in the programming language C) as opposed to Agave, written in Rust. 

There are two versions of the client out there right now: Frankendancer, which is a slightly more performant version of the Agave client, and Firedancer, which is the whole hog. Frankendancer is coming out first and is built through replacing the Agave client piece by piece, like Frankenstein’s monster.

The high-level updates to the Firedancer project are that Frankendancer is live on mainnet, and Firedancer is live on testnet. Someone related to the project told me that another Firedancer update is coming tomorrow, but they wouldn’t say anything further. 

Solana co-founder Anatoly Yakovenko has said that once Firedancer goes to mainnet, he would favor removing the “beta” qualifier from Solana’s “mainnet-beta” status. 

Having multiple validator clients is important for a blockchain’s decentralization (Jito technically runs a separate validator client, but it’s very similar to the Agave client). With just one client, a bug in the code could threaten Solana’s live-ness. With multiple clients, the second client can step in and keep the blockchain running if the first is having issues. I’ve also heard it argued that with just one client, validators don’t have much of an option to opt out of Agave design choices unless they fork the blockchain, which would be a pretty extreme move. Helius CEO Mert Mumtaz and I spoke about this at greater length on the Lightspeed podcast here

Besides decentralization, Firedancer is widely said to be much more performant than the Agave client. The client’s developers have spoken about hitting 1 million transactions per second (TPS) in a test environment. Solana usually handles around 3,000 TPS, and fewer if you only count non-vote transactions that involve users transacting, according to Solana Compass. 

Solana Foundation executive director Dan Albert said there were “audible gasps” when a room of validators were shown Firedancer’s validator monitor GUI. If I was a betting man, I’d put a few bucks down that tomorrow’s update will have to do with Firedancer’s performance. 

The client still has some time to go before reaching mainnet, and even then, users may not actually notice a difference in the blockchain, Lulo co-founder Jesse Brauner told me in a text.

Transactions “always happen fast and [users] won’t notice a millisecond level improvement,” Brauner wrote. 

But either way, Firedancer is moving from idea to reality, and that’s always fun to see.


Get the news in your inbox. Explore Blockworks newsletters:

Tags

Decoding crypto and the markets. Daily, with Byron Gilliam.

Upcoming Events

Javits Center North | 445 11th Ave

Tues - Thurs, March 24 - 26, 2026

Blockworks’ Digital Asset Summit (DAS) will feature conversations between the builders, allocators, and legislators who will shape the trajectory of the digital asset ecosystem in the US and abroad.

recent research

Research Report Templates (27).png

Research

Solana's spot trading landscape will remain bifurcated: prop AMMs will own the short-tail of highly liquid pairs, while passive AMMs continue drifting toward the long-tail. Both can win via vertical integration, but in opposite directions: passive AMMs are moving closer to users through token issuance platforms (e.g., Pump-PumpSwap, MetaDAO-Futarchy AMM), while prop AMMs are moving down the stack into transaction landing services and infrastructure (e.g., HumidiFi-Nozomi). The venues most at risk are legacy AMMs with limited end-user control and no durable, launch-driven source of order flow.

article-image

Some systems improve by failing — and crypto has no choice

article-image

Yield Basis introduces an IL-free AMM design that already dominates BTC DEX liquidity

article-image

Maybe tokenholders don’t need the rights that corporate shareholders have come to expect

article-image

As Hyperliquid and Lighter battle for perps DEX dominance, Boros could capture the structural upside

article-image

Investors are often right about the future, but wrong about the returns

article-image

A look back at 2025, reflections on our industry, and what it means for Blockworks in 2026