DeepSeek releases R1 inference model: MIT open source benchmark OpenAI o1, triggering global AI shock
On January 20, 2025, DeepSeek open sourced the DeepSeek-R1 inference model under the MIT protocol. It was initialized from V3-Base and trained with reinforcement learning-driven training. The mathematical and programming reasoning was benchmarked against OpenAI o1 and is available for free. The application quickly topped the US App Store and caused a shock in the market.
On January 20, 2025, the 
DeepSeek open sourced the DeepSeek-R1 inference model under the MIT license, becoming the most influential release in the DeepSeek series.
Picture: DeepSeek official website homepage and dialogue interface. The DeepSeek application based on R1 will be available for free on iOS and Android in January 2025. It will top the free list of the US App Store on January 27. Related research will be published in the journal Nature in September 2025.
Version overview
| Project | Content |
|---|
Reviews