Hugging Face, Inc.
Company History
Hugging Face History: Founding, Growth & Key Milestones
Founded 2016 in New York, New York
When was Hugging Face founded?
Hugging Face was founded in 2016 in New York, New York by Clément Delangue, Julien Chaumond and Thomas Wolf, making it 10 years old.
- Founded
- 2016
- Founding place
- New York, New York
- Founders
- Clément Delangue, Julien Chaumond and Thomas Wolf
- Headquarters today
- New York, New York, United States
- Milestones documented
- 9 (2016–2026)
- Current CEO
- Clément Delangue
- Acquisitions
- 4
| Year | Event |
|---|---|
| 2016 | Hugging Face Founded |
| 2018 | PyTorch BERT Released |
| 2019 | $15M Series A Led by Lux Capital |
| 2021 | $40M Series B and Gradio Deal |
| 2022 | $100M Series C at $2B; BLOOM Released |
| 2023 | $235M Series D at $4.5B |
| 2024 | 1 Million Models; Argilla and XetHub Acquired |
| 2025 | Pollen Robotics Acquisition and Reachy Mini |
| 2026 | 3 Million Models; Nvidia Agrees to Buy for $12.93B |
- Where was Hugging Face founded?
- Hugging Face was founded in New York, New York in 2016.
- Who founded Hugging Face?
- Hugging Face was founded by Clément Delangue, Julien Chaumond and Thomas Wolf.
- When did Hugging Face start?
- Hugging Face was founded in 2016.
- How old is Hugging Face?
- Hugging Face is 10 years old as of 2026.
Founding Story
Delangue, Chaumond, and Wolf launched Hugging Face in 2016 as an iPhone chatbot for teenagers, a friendly AI you could text, backed by a $1.2 million seed round from Betaworks, SV Angel, and Kevin Durant, among others. To make the bot work they built natural language tools and began open-sourcing them. When Google released BERT in TensorFlow in late 2018, Thomas Wolf published a PyTorch version, pytorch-pretrained-bert. It spread quickly through the research community, was renamed pytorch-transformers and then Transformers, and by 2019 the founders had dropped the app to focus on open-source machine learning.
Founders
Clément Delangue
Clément Delangue is co-founder and CEO of Hugging Face. He led the company from its 2016 teen chatbot app to the open-source pivot and became one of the most visible advocates for open AI, including testimony to the US House Science Committee in June 2023. He raised the company's Series A through D rounds and, in summer 2026, approached Nvidia about a sale, leading to the $12.93 billion agreement announced in September 2026. Clément Delangue was born in 1989.
Julien Chaumond
Julien Chaumond is co-founder and CTO of Hugging Face. He leads the engineering behind the Hub, the platform that stores and serves millions of model repositories and billions of downloads, and its developer-facing products. (Exact birth year remains publicly undisclosed).
Thomas Wolf
Thomas Wolf is co-founder and Chief Science Officer of Hugging Face. He wrote the 2018 PyTorch port of BERT that became the Transformers library, co-led the BigScience workshop that produced the 176B-parameter BLOOM model in 2022, and oversees the company's open science work, including its small model and robotics research. (Exact birth year remains publicly undisclosed).
Key Milestones
Hugging Face Founded
Clément Delangue, Julien Chaumond, and Thomas Wolf launched a chatbot app for teenagers, backed by a $1.2M seed round.
Source: en.wikipedia.orgPyTorch BERT Released
Thomas Wolf open-sourced pytorch-pretrained-bert, the start of the Transformers library.
Source: en.wikipedia.org$15M Series A Led by Lux Capital
Funding to build open-source ML tools after dropping the chatbot app.
Source: en.wikipedia.org$40M Series B and Gradio Deal
Series B led by Addition; acquired Gradio and launched Spaces.
Source: en.wikipedia.org$100M Series C at $2B; BLOOM Released
Round led by Lux Capital; BigScience released the 176B-parameter BLOOM model.
Source: en.wikipedia.org$235M Series D at $4.5B
Backed by Salesforce Ventures, Google, Amazon, Nvidia, AMD, Intel, Qualcomm, IBM, and Sound Ventures.
Source: en.wikipedia.org1 Million Models; Argilla and XetHub Acquired
The Hub passed 1 million public models; Hugging Face bought Argilla and XetHub. A Spaces secrets breach was disclosed in June.
Source: en.wikipedia.orgPollen Robotics Acquisition and Reachy Mini
Bought Pollen Robotics in April and launched the Reachy Mini robot in July; users, models, and datasets nearly doubled during the year.
Source: huggingface.co3 Million Models; Nvidia Agrees to Buy for $12.93B
The Hub passed 3 million models in August; Nvidia signed a definitive acquisition agreement on September 2, expected to close in H1 2027.
Source: finance.yahoo.comMajor Acquisitions
Gradio
2021Acquired the open-source Python library for building web interfaces for machine learning models.
Argilla
2024Acquired the Spanish open-source data curation and annotation platform.
XetHub
2024Acquired the Seattle startup's chunk-based storage technology for very large files and repositories.
Pollen Robotics
2025Acquired the French maker of the open-source Reachy humanoid robot, Hugging Face's first move into hardware.
Keep researching Hugging Face
Hugging Face Company History FAQ
What are the historic milestones in the rapid rise of Hugging Face from 2016 to today?
Key milestones: founded as a chatbot startup (2016), released the PyTorch BERT port that became Transformers (2018), Series A led by Lux Capital (2019), Series B and Gradio acquisition (2021), Series C at $2B and BLOOM (2022), Series D at $4.5B (2023), 1 million models plus Argilla and XetHub deals (2024), Pollen Robotics acquisition (2025), 3 million models and Nvidia's $12.93B acquisition agreement (2026).
How did the release of Google's BERT paper in 2018 change the trajectory of Hugging Face?
When Google released BERT in 2018 with TensorFlow code only, Hugging Face's Thomas Wolf quickly built and open-sourced a clean PyTorch port. The repository gained thousands of GitHub stars overnight, sparking the creation of the universal Transformers library.
What role did Hugging Face play during the generative AI explosion of 2022 to 2024?
Hugging Face served as the central global infrastructure for the open-source generative AI revolution, hosting and serving well-known open weights including Stable Diffusion, Meta's Llama series, Mistral AI, Falcon, Whisper, and DeepSeek, enabling millions of independent AI applications.
What was the landmark BigScience research collaboration and the BLOOM model launch in 2022?
In 2021 to 2022, Hugging Face orchestrated BigScience, a year-long international collaborative effort involving 1,000+ researchers to train BLOOM, a 176B parameter multilingual model trained in 46 natural languages and 13 programming languages, released under a novel Responsible AI License (RAIL).
When did the Hugging Face Hub surpass 1 million open-source models and datasets?
In 2024, the Hugging Face Hub officially crossed the historic milestone of hosting over 1,000,000 public and private machine learning models and hundreds of thousands of datasets, cementing its position as the largest AI repository in human history.
How did Hugging Face expand from Natural Language Processing into Computer Vision, Audio, and Multimodal AI?
Beginning in 2021, Hugging Face expanded Transformers beyond NLP by integrating Vision Transformers (ViT), OpenAI's Whisper audio transcription models, Stable Diffusion image pipelines, and multimodal vision-language models (like LLaVA and Idefics).
What happened to the Open LLM Leaderboard?
Hugging Face launched the Open LLM Leaderboard in 2023 to rank open language models on standard benchmarks. It became one of the most watched rankings in open AI, and Hugging Face retired it in 2025, saying the benchmarks had saturated and newer evaluation approaches were needed.
When did Nvidia agree to buy Hugging Face?
Nvidia signed a definitive agreement on September 2, 2026 and announced it the next day. The $12.93 billion deal is expected to close in the first half of 2027, subject to regulatory approvals.
How did Hugging Face start?
Hugging Face was founded in 2016 in New York by three French entrepreneurs: Clément Delangue, Julien Chaumond, and Thomas Wolf. Hilariously, they did not intend to build enterprise AI infrastructure. They originally launched a quirky, tamagotchi-style iOS chatbot for teenagers. However, the underlying open-source machine learning models they built for the bot became massively popular, forcing them to pivot entirely.
Related Companies
Compare Hugging Face, Inc. History With Competitors
Related Strategic History
Compare founding milestones and corporate evolution
Explore Hugging Face
Core profile pages, founders and related research hubs for this company.