- Taalas, a Toronto-based AI chip development startup, emerges from stealth mode with $50 million funding.
- Funding was secured across two rounds, led by Quiet Capital and Pierre Lamond.
- The company focuses on creating specialized chips for AI models, boasting a cost-efficient approach.
- CEO Ljubisa Bajic, ex-founder of Tenstorrent, heads Taalas along with early Tenstorrent engineers.
- Taalas aims to reset AI cost structure, enable significant model size growth, and facilitate local AI model processing.
- Lamond’s involvement signals industry confidence; Taalas founders possess extensive chip development experience.
Main AI News:
Toronto-based Taalas, a pioneering artificial intelligence (AI) chip development startup, emerges from stealth mode with an impressive $50 million secured in funding. The capital infusion, orchestrated across two strategic funding rounds, was led by Quiet Capital and Eclipse Ventures advisor Pierre Lamond.
Taalas strategically closed a $12-million fundraise in September 2023, followed by a recent funding surge of $38 million in the preceding month, as reported by The Information. The startup, established in 2023, is dedicated to revolutionizing the AI landscape through the development of an automated flow for rapid implementation of specialized chips. These chips boast the capability of individually running diverse AI models, with one key differentiator being the capacity to house an entire large AI model without external memory requirements. Leveraging hard-wired computation, Taalas aims to position a single chip to outperform small GPU data centers, presenting a potential paradigm shift in the cost dynamics of AI hardware development.
Matt Humphrey, Partner at Quiet Capital, emphasizes the groundbreaking nature of Taalas’ ‘direct-to-silicon’ foundry, stating, “We believe the Taalas’ direct-to-silicon’ foundry unlocks three fundamental breakthroughs: dramatically resetting the cost structure of AI today, viably enabling the next 10 to 100x growth in model size, and efficiently running powerful models locally on any consumer device.” This, he asserts, is a pivotal mission for the future scalability of AI in computing.
Founded by CEO Ljubisa Bajic, a notable figure in the AI sector and founder of Tenstorrent, Taalas boasts a team including Drago Ignjatovic and Lejla Bajic, early engineers at Tenstorrent. Tenstorrent itself, a company specializing in the design of computers for AI model training and execution, secured a substantial $100 million USD in funding in August 2023 from investors such as Samsung. Ljubisa Bajic, the former CEO of Tenstorrent, transitioned to helm Taalas in September 2023.
Pierre Lamond, an advisor at Eclipse Ventures, played a pivotal role as an independent investor in Taalas’ funding rounds. Eclipse Ventures, with Lamond at its helm, had previously supported Tenstorrent in its August fundraising endeavors. Lamond emphasizes the difficulty and risk inherent in chip development, praising Taalas’ founders for their extensive experience, ingenuity, and persistence. “Taalas’ founders have produced numerous cutting-edge chips and systems. Their track record in the industry is second to none,” he stated confidently.
Conclusion:
Taalas’ emergence with $50 million funding marks a significant stride in AI hardware innovation. The company’s focus on specialized chips reflects a pivotal shift in the industry towards cost-effective and scalable AI solutions. With experienced leadership and notable investor backing, Taalas is poised to drive substantial advancements in AI hardware technology, potentially reshaping the landscape of AI development and deployment.
