DeepSeek R1 is an LLM that excels in reasoning tasks, particularly in mathematics, coding, and complex problem-solving. It leverages reinforcement learning and a chain-of-thought approach to generate structured, interpretable responses.

Some other noteworthy features of DeepSeek R1 include its ability to provide both reasoning processes and final answers, and its cost-efficiency compared to similar models.

Metric	Value
Parameter Count	685 billion
Mixture of Experts	Yes
Active Parameter Count	37 billion
Context Length	128,000 tokens
Multilingual	Yes
Quantized*	Yes
Quantization*	FP8

*Quantization is specific to the inference provider and the model may be offered with different quantization levels by other providers.