DeepSeek, a Chinese startup, has launched an updated version of its R1 reasoning AI model on Hugging Face. This release, shared via WeChat, is described as a "minor" upgrade and is available under a permissive MIT license, allowing for commercial usage. Weighing in at 685 billion parameters, the model's complexity suggests that it requires high-end hardware for operation. Having gained attention for its performance earlier this year, DeepSeek faces scrutiny from regulators who express concerns regarding potential national security risks associated with its technology.
The updated R1 model from DeepSeek, boasting 685 billion parameters, represents a notable advancement in AI technology that could challenge established leaders like OpenAI.
DeepSeek's R1 model is available under an MIT license, enabling commercial use, though the lack of detailed descriptions implies limited accessibility for everyday developers.
Collection
[
|
...
]