Hugging Face Serves Nearly 1 Exabyte of Open AI Models Monthly
Hugging Face announced it now distributes nearly 1 exabyte of open-source machine learning models and datasets globally each month, necessitating the construction of its own Content Delivery Network. The platform also reached 20 million active AI builders and hosts 300 petabytes of content, including a weekly addition of 9 petabytes. October marked the company's record-breaking month for users and revenue, highlighting the exponential growth and infrastructure demands of the open-source AI ecosystem.
Editorial responsibility
- No named human review is recorded for this page.
- Reports are grouped by semantic similarity and deterministic rules. Language models may assist titles, summaries, translation and cross-source analysis; the page reads the event directly, while its address stays stable when the title changes.
- Summary covers the current reports
Cross-source coverage
Reporting timeline
Hugging Face Serves Nearly 1 Exabyte of Open Models Monthly
Julien Chaumond, representing Hugging Face, announced that the platform now serves nearly 1 exabyte of open models and datasets over the internet each month. To manage this massive scale of data distribution, the organization has been compelled to build its own Content Delivery Network (CDN). This infrastructure development highlights the exponential growth in the consumption of open-source artificial intelligence resources and underscores the technical challenges associated with global model distribution at such volumes.
Read sourceHugging Face Distributes Nearly 1 Exabyte of AI Models Monthly
Clement Delangue, a representative from Hugging Face, announced that the platform is currently distributing approximately 1 exabyte (equivalent to 1 billion gigabytes) of machine learning models and datasets globally every month. The statement highlights the massive scale at which open intelligence is being shared with the world. This metric underscores the rapid growth and accessibility of artificial intelligence resources provided by the organization. The post characterizes this distribution as a significant achievement in making open intelligence available at scale, emphasizing the global reach of these digital assets.
RT @ClementDelangue: We just crossed: - 20M AI builders on @huggingface - 300PB of models, datasets, agent traces, and RL environments hos…
Clement Delangue announced that the Hugging Face platform has reached two significant milestones: it now hosts over 20 million AI builders and contains 300 petabytes of data. This data includes a diverse range of assets such as machine learning models, datasets, agent traces, and reinforcement learning environments. The announcement highlights the rapid growth of the open-source AI community and the expanding scale of resources available for developers to build and train artificial intelligence systems. The figures represent a concrete metric of adoption and infrastructure capacity within the current AI ecosystem.
Show 1 older updatesHide older updates
Hugging Face Surpasses 20M AI Builders and Hosts 300PB of Data
Clement Delangue announced that Hugging Face has reached two major milestones: the platform now hosts over 20 million AI builders and contains 300 petabytes of models, datasets, agent traces, and reinforcement learning environments. Notably, users added 9 petabytes of content in just the last week alone. The announcement also highlighted that October was the company's biggest month ever regarding daily active users (DAUs), monthly active users (MAUs), and revenue. These figures underscore the rapid growth and increasing adoption of open-source artificial intelligence tools within the developer community.
Read source