Insider Brief
- Taalas Inc., an AI and silicon innovation company, has emerged from stealth mode with $50 million in funding to focus on transforming AI’s computational efficiency by directly implementing deep learning models into silicon, aiming for a 1000x improvement.
- The company’s proprietary technology allows for the storage of entire large AI models on a single chip without external memory, significantly reducing AI’s cost structure and potentially outperforming small GPU data centers.
- Set to tape out its first large language model chip in Q3 2024 for early customer access in Q1 2025, Taalas is led by a team with a strong track record in the semiconductor industry, including experience from Tenstorrent, AMD, and NVIDIA.
PRESS RELEASE — TORONTO/March 5, 2024 /PRNewswire/ — Taalas Inc., an innovator in AI and silicon, is pleased to announce today that it has exited stealth mode and raised $50 million dollars over two rounds of funding led by Pierre Lamond and Quiet Capital.
Over the last year, AI has undergone a large scale productization and has already begun reshaping the world. Concurrently, deep learning models have become the world’s most demanding computational workload, unsustainably capital intensive, power hungry, and GPU constrained.
“Artificial intelligence is like electrical power — an essential good that will need to be made available to all. Commoditizing AI requires a 1000x improvement in computational power and efficiency, a goal that is unattainable via the current incremental approaches. The path forward is to realize 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,” said Ljubisa Bajic, Taalas’ CEO.
Taalas is developing an automated flow for rapidly implementing all types of deep learning models (Transformers, SSMs, Diffusers, MoEs, etc.) in silicon. Proprietary innovations enable one of its chips to hold an entire large AI model without requiring external memory. The efficiency of hard-wired computation enables a single chip to outperform a small GPU data center, opening the way to a 1000x improvement in the cost of AI.
“We believe the Taalas ‘direct-to-silicon’ foundry unlocks three fundamental breakthroughs: dramatically resetting the cost structure of AI today, viably enabling the next 10–100x growth in model size, and efficiently running powerful models locally on any consumer device. This is perhaps the most important mission in computing today for the future scalability of AI. And we are proud to support this remarkable n-of-1 team as they do it,” said Matt Humphrey, Partner at Quiet Capital.
The company is taping out its first large language model chip in the third quarter of 2024 and planning to make it available to early customers in the first quarter of 2025.
“Making chips is a difficult and risky endeavour in which success requires experience, ingenuity, and persistence. Taalas’ founders have produced numerous cutting-edge chips and systems. Their track record in the industry is second to none,” said Pierre Lamond, a legend in the field, whose work across Fairchild, National Semiconductor, Sequoia, and many other organizations, helped lay the foundation of the semiconductor industry.
Taalas was founded by Ljubisa Bajic, Drago Ignjatovic, and Lejla Bajic. Prior to co-founding Taalas, Ljubisa founded Tenstorrent in 2016. Drago and Lejla joined Tenstorrent soon after as early engineering leaders. The team has spent decades collectively working together on a long list of AI processors, GPUs, and CPUs across Tenstorrent, AMD, and NVIDIA.
