Skip to content
GenAI BR
Open modelsIn the archive since 06 Sept 2026

DeepSeek-R1: pure reinforcement learning, with no human reasoning examples, taught the model to verify itself

Open modelstechnicaltechnicalpaper · DeepSeek · Nature 645 (2025) · 22 Jan 2025
Download card
What it is
R1-Zero was trained only with reinforcement learning on verifiable rewards (correct answers in math and code), without supervised reasoning data; R1 adds a cold-start data stage and further training. The work was published in Nature in 2025, and the weights are open.
What was demonstrated
Self-reflection and verification behaviors emerged from reinforcement; versions distilled to small models (1.5B to 70B) transferred part of the capability.
What was not
R1-Zero mixed languages and was barely readable; the recipe requires verifiable rewards, which limits the method to domains with a right answer. Inference cost of long reasoning is high.
Why it matters
It defined the open “reasoning model” recipe that China has been iterating on ever since.

Same theme

Open models
  1. Tencent open-sourced a 770B MoE under Apache 2.0 — and it edges Kimi K3 on code with a quarter of the parameters

    Open modelstechnicaltechnical, decision-makerannouncement · Tencent · 28 Aug 2026
  2. GLM-5.3-Flash: an MIT-licensed 320B model served entirely on Chinese chips, at US$ 0.15 per million tokens

    Infra and costtechnicaltechnical, decision-makerannouncement · Z.ai · 26 Aug 2026
  3. DeepSeek-V3: a 671B model trained in 2.79 million H800 hours, without a single loss spike

    Infra and costtechnicaltechnical, decision-makerpaper · DeepSeek · 27 Dec 2024
  4. Llama 3: the 92-page report that showed how a 405B dense model is trained — and what “open” meant in 2024

    Open modelsintroductorytechnical, beginnerpaper · Meta · 31 Jul 2024

Get the next Cinco

Tuesday, 7am BRT, in your inbox. Five items, with sources. No daily newsletter, no promotions, and unsubscribing is one click.

Just the e-mail. No third-party trackers on this page. Privacy policy.