DeepSeek releases new open-source AI model on Huawei chips
Chinese AI startup DeepSeek has released a new open-source AI model, its first major launch since the viral R1 model. The new model comes in two versions and is built on Huawei's Ascend processors, marking a significant shift away from reliance on Western chip technology. This release underscores China's push for self-sufficiency in AI hardware amid ongoing US export restrictions. The model is open-sourced, continuing DeepSeek's strategy of making its AI technology publicly available.
Key facts
- DeepSeek released a new open-source AI model.
- It is the first major release since R1 went viral.
- The model comes in two versions.
- It is built on Huawei's Ascend processors.
- The release highlights China's self-sufficiency in AI chips.
- The model is open-sourced.
Entities
Institutions
- DeepSeek
- Huawei
Locations
- China
Sources
- Quartz —