Toronto-based artificial intelligence (AI) startup Cohere has secured $125 million USD ($159 million CAD) in Series B financing to fuel its United States expansion plans.
Cohere, which aims to make natural language processing (NLP) more accessible, first emerged from stealth in May 2021 and announced a $40 million USD Series A round in September. Since then, Cohere has opened up general access to its API and inked a partnership to leverage Google Cloud’s AI infrastructure.
The startup plans to use the capital to develop its platform, expand its team, and open a Palo Alto office.
The startup plans to use the fresh capital to fuel its platform development efforts, expand its team, and open a new office in Palo Alto, California, which will be run by former Apple Director Bill MacCartney, who has joined Cohere as VP of engineering and machine learning.
Cohere’s Series B round was led by Tiger Global, with follow-on participation from founding investor Radical Ventures, Index Ventures, and Section 32. The fresh capital brings Cohere’s total funding to over $170 million USD.
Co-founded by CEO Aidan Gomez, Ivan Zhang, and Nick Frosst in 2019, Cohere claims its API enables companies to deploy NLP capabilities across their businesses without requiring supercomputing infrastructure or AI expertise, which it says “radically reduces the cost for companies of all sizes to access leading AI models.”
“Cohere has been able to solve one of the most difficult obstacles to the broad adoption of NLP technology: ease of use,” said MacCartney. “Today, we’re well-positioned for success as we scale up operations and expand our international footprint.”
Prior to joining Cohere, MacCartney served as director of proactive intelligence at Apple, where he led a team of more than 120 scientists and engineers in developing models for predicting user intentions and actions. He also currently works as consulting professor of computer science at Stanford, where he co-developed and has taught a graduate-level NLP course over the past twelve years.
Over the past eight months, Cohere has doubled its headcount. Since closing its Series A round, the company claims its user base has grown by over 800 percent.
“For the first time, we’ve brought to market an effective NLP solution that is practical, accessible, and safe,” said Gomez, Cohere’s CEO. “With the opening of our new Palo Alto office, we’re continuing to scale in all directions, bringing aboard new talent and rapidly increasing our compute capacity to train our next generation large language models.”
In addition to Tiger, Radical, Index, and Section 32, Cohere’s investors include Turing Award winner Geoffrey Hinton, Fei-Fei Li, Pieter Abbeel, and Raquel Urtasun of Toronto-based self-driving startup Waabi.
According to The Globe and Mail, Gomez and Frosst both previously worked with Hinton, and met while working at Google Brain, a deep learning AI research team under Google AI’s umbrella. In 2017, Gomez, Cohere’s CEO, co-authored “Attention is All You Need,” a paper that introduced the “transformer” architecture behind Cohere’s platform.