Taalas Inc., headquartered in Toronto, Canada, has successfully secured $50 million in funding through two rounds led by Pierre Lamond and Quiet Capital, marking a significant milestone in the company’s journey towards revolutionizing artificial intelligence (AI) and silicon integration.

The funds raised are earmarked for expanding operational capacities and intensifying development efforts. Co-founded by Ljubisa Bajic, Drago Ignjatovic, and Lejla Bajic, Taalas specializes in creating an automated workflow for integrating various deep learning models, including Transformers, SSMs, Diffusers, and MoEs, into silicon.

What sets Taalas apart is its proprietary technology that enables an entire large AI model to be housed within a single chip without the need for external memory. The company is currently in the process of taping out its inaugural large language model chip, slated for completion in the third quarter of 2024, with plans to roll it out to early adopters by the first quarter of 2025.

Commenting on this latest development, Ljubisa Bajic emphasized the significance of making AI accessible to all, likening it to electrical power. He stressed the necessity for a 1000x enhancement in computational power and efficiency, which conventional approaches cannot achieve. Bajic advocates for a paradigm shift towards implementing deep learning models directly into silicon, deeming it the most sustainable path towards advancing AI technology.

Under Taalas’s vision, the focus is on creating Hardcore Models, a platform capable of transforming any AI model into custom silicon, boasting efficiency levels that are 1000x greater than their software counterparts. By embodying AI models within native hardware, Taalas aims to revolutionize the landscape of AI computing, making it more accessible and efficient for various applications.

In essence, Taalas’s mission underscores the notion that the model is the computer, advocating for the optimization of AI through hardwired silicon integration. With Hardcore Models, the company aims to bridge the gap between traditional AI software and hardware, ushering in a new era of computational efficiency and accessibility.

For more such insightful news & updates around AI or Automation, explore other articles here. You can also follow us on Twitter.