Quick take:
The company previously raised $20 million from investors including Distributed Global, North Island Ventures, and Delphi Digital.
Nous Research is using the Solana blockchain to train its open-source AI models.
A big percentage of the new funding will go to compute power and expanding its research capabilities.
Nous Research, a decentralised AI startup has raised $50 million in a Series A round almost entirely funded by crypto-focused venture firm Paradigm. The fundraising closed at a token valuation of $1 billion, a source who asked for anonymity told Fortune.
The fundraising follows the company’s $20 million funding in seed rounds backed by the likes of Distributed Global, North Island Ventures, and Delphi Digital.
Nous is building decentralised AI training for open-source AI models on Solana with a view of taking on the centralised AI giants like DeepSeek and OpenAI.
The company was founded by developers who did not originally opt to use blockchain technology but later chose it for its coordination and incentive capabilities, which allow it to leverage the power of a decentralised GPU network.
Unlike in centralised systems where AI models are trained in a central data centre, Nous plans to train its models in a completely distributed way, harnessing spare computing capacity from people around the world, Fortune reported.
“We very much came from a mentality that we want to create and serve the world’s best AI,” said cofounder Karan Malhotra.
According to Malhotra, Nous is leveraging blockchain technology to drive participation. “We think of the incentive mechanism behind crypto to push people to actually utilize their idle compute less as a donation but more as a transaction.”
Malhotra also believes that using blockchain technology to train AI models de-incentivises bad actors in a distributed training approach, making sure that the process is not poisoned by users who send inappropriate data.
Stay on top of things:
Subscribe to our newsletter using this link – we won’t spam!
Follow us on X and Telegram.
