DeepSeek-R1 · GitHub Models · GitHub

DeepSeek-R1 excels at reasoning tasks utilizing a step-by-step training procedure, such as language, clinical thinking, and coding tasks. It includes 671B total parameters with 37B active specifications, and 128k context length.

DeepSeek-R1 constructs on the progress of earlier reasoning-focused models that improved efficiency by extending Chain-of-Thought (CoT) thinking. DeepSeek-R1 takes things further by integrating support learning (RL) with fine-tuning on carefully chosen datasets. It progressed from an earlier variation, DeepSeek-R1-Zero, which relied solely on RL and revealed strong reasoning skills but had issues like hard-to-read outputs and language disparities. To resolve these restrictions, DeepSeek-R1 includes a little quantity of cold-start information and follows a refined training pipeline that blends reasoning-oriented RL with monitored fine-tuning on curated datasets, resulting in a design that attains advanced performance on thinking criteria.

Usage Recommendations

We suggest adhering to the following configurations when utilizing the DeepSeek-R1 series designs, benchmarking, to accomplish the anticipated performance:

– Avoid adding a system prompt; all directions need to be included within the user timely.
– For mathematical problems, it is recommended to consist of a directive in your timely such as: « Please factor action by step, and put your final response within boxed . ».
– When examining model performance, it is advised to perform several tests and average the results.

Additional suggestions

The design’s reasoning output (consisted of within the tags) may include more damaging material than the model’s last reaction. Consider how your application will use or display the reasoning output; you might want to suppress the reasoning output in a production setting.