MIT spin-off Liquid AI has secured major funding to scale its efficient AI models, inspired by the neural architecture of microscopic worms rather than traditional transformer-based systems.
The Boston-based startup has secured $250 million in Series A funding led by AMD, valuing the company at $2.3 billion just a year after its founding.
Key points:
- AMD led the round, joined by OSS Capital, Duke Capital Partners, and PagsGroup.
- Liquid AI’s models, based on the nematode brain, promise greater efficiency with less data and computing power.
- The technology targets industries from financial services to consumer electronics.
Traditional transformer-based models, like those behind ChatGPT, are resource-intensive and expensive to scale. Liquid AI’s “liquid foundation models” claim to do more with less, reducing the dependency on massive data centers and opening up on-device AI possibilities.
Instead of mimicking the human brain like traditional neural networks, the company's technology draws inspiration from the Caenorhabditis elegans, a microscopic worm with a surprisingly efficient neural structure. This tiny creature, measuring just one millimeter in length, has become the blueprint for what the company calls "liquid foundation models."
"We have been proving the technology in the last year, making sure that an alternative structure to transformers can be scaled," said Ramin Hasani, Liquid AI's co-founder and CEO, in an interview with Bloomberg. "This funding will help get us to the next level."
AMD's involvement signals strong interest in this alternative approach. "We just fundamentally believe that there is a lot of innovation continuing in AI and driving models forward," said Mathew Hein, AMD's senior vice president and chief strategy officer of corporate development.
While Liquid AI hasn't named specific clients, they're reportedly working with businesses in consumer electronics, financial services, and other sectors. The company plans to use the fresh capital to scale their computing infrastructure and accelerate product development for edge and on-premise deployments.
The round included participation from OSS Capital, Duke Capital Partners, and PagsGroup, building on the company's previous $46.6 million seed funding. As the AI industry grapples with efficiency challenges, Liquid AI's worm-inspired approach might prove that sometimes, the smallest creatures can inspire the biggest innovations.