DeepSeek-R1 · GitHub Models · GitHub

DeepSeek-R1 stands out at reasoning jobs utilizing a step-by-step training procedure, such as language, scientific reasoning, and coding jobs. It features 671B total specifications with 37B active specifications, and 128k context length.

DeepSeek-R1 builds on the development of earlier reasoning-focused designs that enhanced efficiency by extending Chain-of-Thought (CoT) reasoning. DeepSeek-R1 takes things even more by integrating support knowing (RL) with fine-tuning on thoroughly chosen datasets. It developed from an earlier version, DeepSeek-R1-Zero, which relied exclusively on RL and showed strong thinking abilities but had concerns like hard-to-read outputs and language inconsistencies. To deal with these limitations, DeepSeek-R1 includes a percentage of cold-start data and follows a refined training pipeline that RL with supervised fine-tuning on curated datasets, resulting in a design that achieves cutting edge efficiency on reasoning criteria.

Usage Recommendations

We recommend adhering to the following configurations when making use of the DeepSeek-R1 series models, including benchmarking, to achieve the anticipated efficiency:

– Avoid including a system timely; all guidelines should be contained within the user prompt.
– For mathematical issues, it is advisable to include a regulation in your timely such as: “Please reason step by action, and put your final response within boxed .”.
– When evaluating model performance, it is suggested to carry out multiple tests and average the results.

Additional suggestions

The model’s reasoning output (included within the tags) may contain more damaging material than the design’s last reaction. Consider how your application will utilize or show the thinking output; you might desire to reduce the thinking output in a production setting.