DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

DeepSeek-AI, Daya Guo, Dejian Yang|January 22, 2025arXiv

Key Takeaway

RL alone can teach LLMs to reason step-by-step — DeepSeek-R1 matched o1 performance at a fraction of the training cost and proved that reasoning can be distilled into small models.

Summary

Demonstrates that pure reinforcement learning without supervised fine-tuning can produce strong reasoning capabilities. DeepSeek-R1 matches OpenAI o1 on math and coding benchmarks, and distilled versions (1.5B-70B) retain most reasoning ability.

reasoning training scaling