arcflare

deepseek-r1

7BReasoning Official

by DeepSeek

DeepSeek-R1 — an open reasoning model trained with reinforcement learning, with strong math and coding.

reasoningchain-of-thoughtmathMIT
arcflare run deepseek-r1

Quickstart

arcflare — zsh
$ 

DeepSeek-R1

An open reasoning model trained with large-scale reinforcement learning. It

"thinks" step by step before answering, with strong performance on math, code,

and logic. The distilled sizes run on consumer hardware.

arcflare run deepseek-r1
R1 weights are MIT-licensed. Distilled variants inherit their base model's license.