AI-to-silicon startup Taalas raises $50m in two rounds
Toronto-based Taalas has secured funding to develop a platform for turning AI models into custom silicon.
The startup was founded in August 2023 by Tenstorrent founder Ljubisa Bajic – a former architect at both AMD and Nvidia. It is developing an automated flow for implementing all types of deep learning models (Transformers, SSMs, Diffusers, MoEs) directly into silicon. It says this process will enable a single chip to outperform a small GPU data centre, and could deliver a 1000x reduction in the cost of AI.
Now, with its new funding and out of stealth mode, Taalas plans to unveil its first LLM chip in the third quarter of 2024, and commercially available in Q1 2025.
"Artificial intelligence is like electrical power – an essential good that will need to be made available to all,” said Ljubisa Bajic, Taalas CEO. "Commoditising AI requires a 1,000x improvement in computational power and efficiency, a goal that is unattainable via the current incremental approaches. The path forward is to realise that we should not be simulating intelligence on general purpose computers, but casting intelligence directly into silicon. Implementing deep learning models in silicon is the straightest path to sustainable AI.”
