DeepSeek Updates R1 Reasoning AI Model, Releases on Hugging Face
DeepSeek, a Chinese AI startup, has released an updated version of its R1 reasoning AI model on the Hugging Face developer platform. The announcement was made via a WeChat message on Wednesday morning, marking another significant step for the company in the competitive AI landscape.
The updated R1 model is available under the permissive MIT license, allowing for commercial use. According to DeepSeek’s WeChat announcement, this release is considered a “minor” upgrade to the original R1. The Hugging Face repository includes configuration files and weights, which are the internal components that govern the model’s behavior, but lacks a detailed description of the specific changes implemented.
With 685 billion parameters, the R1 model is substantial in size. The term “parameters” is synonymous with “weights” in AI terminology. Due to its large scale, the model is unlikely to run on standard consumer hardware without modifications.
DeepSeek gained prominence earlier this year with the initial release of R1, which rivaled models from OpenAI. The company has also faced scrutiny from regulators in the United States, who have raised concerns about potential national security risks associated with DeepSeek’s technology.
