Dataset
DeepSeek-R1 model weights
Weights for DeepSeek-R1 from Huggingface We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With...
About this listing
- Read from
- Academic Torrents
- Licence, as the source states it
-
MIT License
Read the licence text - The source's own words
- https://i.imgur.com/q6NKD6T.png ## License This code repository and the model weights are licensed under the MIT License.
This listing was read from the catalogue of Academic Torrents. The material itself is not held here.
This page is a pointer. It carries the title, the description, the source and the licence of an item that another library publishes, and it links you to that library. Nothing on this page lets you obtain the item from here.
If you believe a listing points at material that should not be linked, see Copyright & Takedowns. A valid report removes the link.
Quoted text from a source is shortened where it names a way to obtain a file.