DeepSeek-R1 · GitHub Models · GitHub

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

DeepSeek-R1 constructs on the progress of earlier reasoning-focused models that improved 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 progressed from an earlier variation, DeepSeek-R1-Zero, which relied exclusively on RL and showed strong reasoning skills however had concerns like hard-to-read outputs and language disparities. To attend to these constraints, DeepSeek-R1 integrates a little quantity of cold-start data and follows a refined training pipeline that blends reasoning-oriented RL with supervised fine-tuning on curated datasets, resulting in a design that achieves cutting edge efficiency on thinking criteria.

Usage Recommendations

We suggest adhering to the following setups when utilizing the DeepSeek-R1 series designs, including benchmarking, to achieve the anticipated efficiency:

– Avoid adding a system prompt; all instructions should be included within the user prompt.
– For mathematical problems, it is suggested to consist of a regulation in your prompt such as: “Please factor action by step, and put your last response within boxed .”.
– When examining design performance, it is advised to conduct numerous tests and balance the outcomes.

Additional recommendations

The model’s thinking output (consisted of within the tags) might include more hazardous content than the model’s last action. Consider how your application will use or display the thinking output; you might desire to suppress the thinking output in a production setting.