Tech giants like Google, Amazon, and Nvidia boost AI startup Hugging Face to a $4.5 billion valuation with their investments.

Hugging Face, a New York-based AI firm, secured $235 million in funding, valuing the company at $4.5 billion, with support from major players in the tech industry, including Google, Amazon, Nvidia, Salesforce, AMD, Intel, IBM, and Qualcomm. The CEO of Hugging Face, Clement Delangue, explained that these funds will primarily be allocated to attracting top talent, enhancing competitiveness in the artificial intelligence sector.

Startups specializing in AI models have seen their valuations soar as both large corporations and venture capitalists look to invest in the recent AI surge, sparked by the release of Microsoft-backed OpenAI’s ChatGPT chatbot last year.

Hugging Face’s substantial valuation and its list of prominent backers underscore the increasing trend of collaborative AI development, particularly following Meta’s release of the Llama large language model, which is freely available to most companies.

Unlike some highly valued AI startups like OpenAI or Cohere, which keep their technology as trade secrets and charge customers for access through APIs, Hugging Face provides a platform for AI developers to share code, models, data sets, and utilize the company’s developer tools for simplifying the deployment of open-source AI models. It also hosts essential components of modern AI models known as weights, which consist of lists of numbers.

While Hugging Face has developed some models, such as BLOOM, its primary offering is a web platform where users can upload models and their associated weights. Additionally, it offers a suite of software tools, called libraries, to expedite model deployment, manage extensive datasets, and assess performance. The platform also hosts certain AI models in a web interface for end users to experiment with.

This approach is akin to GitHub, a code repository that coders worldwide use to share their projects during development, and Hugging Face believes that most AI companies will require tools to create their own models and technology. The company’s co-founder and CEO, Clement Delangue, anticipates that AI developers will rely on Hugging Face extensively to carry out their work, given its current popularity among AI builders.

One compelling reason for significant corporate investments in Hugging Face is the active use of the platform by their employees, Delangue noted. He envisions a significant increase in the number of software developers working with AI models in the coming years, potentially reaching 100 million.

Although much attention has recently focused on large language models like ChatGPT and Llama, which specialize in generating text, Hugging Face accommodates various AI models, including those that generate music or images, translate languages, or identify objects within images. The company hosts a staggering 500,000 different AI models, 250,000 datasets, and boasts 10,000 paying customers.

As for the origin of the company’s name, “Hugging Face” is inspired by the hugging face emoji – a smiling face enclosed by two open hands. This name and logo stem from the company’s inception, which began as an iPhone chatbot app. However, as Hugging Face open-sourced some of its machine-learning code, it gained traction among AI developers, prompting the company to shift its focus. Delangue humorously remarked that they once joked about becoming the first company to go public with an emoji instead of a traditional three-letter ticker symbol, suggesting they might even approach Nasdaq with this unconventional idea during this funding round.