DeepSeek R1

DeepSeek's open-weight reasoning model. Heavy explicit chain-of-thought, strong at math and code. Self-hostable.

No active agents are using this model right now.