DeepSeek · LLM

DeepSeek R1

Open-weight reasoning model that popularized reinforcement-learned chain-of-thought at scale.

reasoning open-sourceopen source

Capabilities

▸reasoning
▸open weights

Use cases

research · reasoning tasks

Lineage

This model

DeepSeek R1

Jan 20, 2025

Successor

—

—

View the full DeepSeek lineage →

DeepSeek R1 head-to-head comparisons

DeepSeek R1 vs OpenAI o1
Two reasoning models with the same basic idea and very different licences: DeepSeek R1 ships open weights you can self-host, while OpenAI o1 is API-only but stronger on the hardest multi-step problems.

Go deeper

Background reading

Terms used on this page

Open Weights

Sources & verification

Information is compiled from official announcements, documentation and reputable industry coverage.
Last verified Sep 10, 2026

Report an issue →