# DeepSeek-R1-Zero

> The RL-only reasoning model that learned to think with no supervised fine-tuning at all.

- Maker: DeepSeek
- Line: DeepSeek R1
- Canonical page: https://ai-tldr.dev/models/deepseek-r1-zero/
- Updated: 2026-06-18

## In short

DeepSeek-R1-Zero is the first model in DeepSeek's R1 reasoning line, released on January 20, 2025. It has a 128K-token context window. Its weights can be downloaded under the MIT licence.

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Context window | 128K tokens (max output 32,768 tokens) | [1] |
| Licence | MIT · open weights | [1] |
| Released | 20 January 2025 | [2] |
| Parameters | 671B total / 37B activated (Mixture-of-Experts) | [1] |
| Knowledge cutoff | Inherited from DeepSeek-V3-Base (training data up to ~July 2024) | [1] |

## Sources

1. [DeepSeek-R1-Zero on Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero)
2. [DeepSeek-R1 GitHub repository](https://github.com/deepseek-ai/DeepSeek-R1)

## Overview

DeepSeek-R1-Zero is the first model in DeepSeek's R1 reasoning line, released on January 20, 2025. It is the headline science experiment behind the R1 launch: starting from the DeepSeek-V3-Base mixture-of-experts model (671B total parameters, 37B activated), DeepSeek trained it using large-scale reinforcement learning alone, with no supervised fine-tuning step beforehand. The reward was simple, mostly correctness on math and code, and the model was left to figure out how to reason on its own.

What makes DeepSeek-R1-Zero notable is what emerged from that pure-RL process. Without ever being shown human-written reasoning traces, the model spontaneously developed long chains of thought, self-verification, reflection, and an 'aha moment' where it learned to reallocate thinking time to harder steps. Its AIME 2024 score climbed from 15.6% to 71.0% pass@1 over training (86.7% with majority voting), matching OpenAI's o1-0912 on that benchmark. This was the first open research to publicly validate that reasoning capability can be incentivized through reinforcement learning alone.

The catch is that DeepSeek-R1-Zero is hard to use in practice. The paper openly documents that its outputs suffer from poor readability and language mixing (switching between English and Chinese mid-thought), with chaotic formatting. Those flaws are exactly why DeepSeek built DeepSeek-R1 on top of it, adding a cold-start dataset and multi-stage training to clean up the output while keeping the reasoning gains. R1-Zero remains open under the MIT license on Hugging Face mainly as a research baseline rather than a polished assistant. The R1 work was later peer-reviewed and published on the cover of Nature (Vol. 645, 18 September 2025).

## Benchmarks

- AIME 2024 (pass@1): 71% ([source](https://arxiv.org/html/2501.12948v1))
- AIME 2024 (cons@64, majority voting): 86.7% ([source](https://arxiv.org/html/2501.12948v1))
- MATH-500 (pass@1): 95.9% ([source](https://arxiv.org/html/2501.12948v1))
- GPQA Diamond (pass@1): 73.3% ([source](https://arxiv.org/html/2501.12948v1))
- LiveCodeBench (pass@1): 50% ([source](https://arxiv.org/html/2501.12948v1))

## Strengths

- Proves reasoning can emerge from reinforcement learning alone, with zero supervised fine-tuning
- Strong math performance: 71.0% pass@1 on AIME 2024 (86.7% with majority voting), matching OpenAI o1-0912 on AIME
- Open weights under the permissive MIT license, free for commercial and derivative use
- Efficient MoE design: only 37B of 671B parameters active per token
- Exposes its full chain-of-thought, useful for studying how reasoning behaviors self-organize

## Best for

- Research baseline for studying RL-driven reasoning and self-evolution in LLMs
- A starting point for teams experimenting with pure-RL or GRPO training pipelines
- Math and competition-style problem solving where readable formatting is not required
- Ablation comparisons against DeepSeek-R1 to measure the value of cold-start SFT data
- Generating reasoning traces for distillation or analysis (with post-processing for readability)

## DeepSeek R1 versions

- [DeepSeek-R1-0528](https://ai-tldr.dev/models/deepseek-r1-0528/) (28 May 2025)
- [DeepSeek-R1](https://ai-tldr.dev/models/deepseek-r1/) (20 January 2025)
- DeepSeek-R1-Zero (20 January 2025) — this page

## FAQ

### What is the difference between DeepSeek-R1-Zero and DeepSeek-R1?

DeepSeek-R1-Zero is trained with reinforcement learning only, directly on DeepSeek-V3-Base, with no supervised fine-tuning. It reasons well but its output suffers from poor readability and language mixing. DeepSeek-R1 adds a cold-start dataset and multi-stage training on top of that RL recipe to fix the readability problems while keeping the reasoning gains, which is why R1 is the model people actually deploy.

### How was DeepSeek-R1-Zero trained?

It was trained purely with large-scale reinforcement learning using Group Relative Policy Optimization (GRPO), a critic-free RL algorithm, starting from DeepSeek-V3-Base. There was no supervised fine-tuning step. Reasoning behaviors like long chain-of-thought, self-verification, and reflection emerged on their own during RL rather than being taught from human examples.

### Can I use DeepSeek-R1-Zero via an API?

Not as a dedicated product. DeepSeek published API pricing only for DeepSeek-R1, not for R1-Zero. R1-Zero is released as open weights under the MIT license on Hugging Face, so you can download and self-host it, but DeepSeek never offered it as a paid hosted endpoint.

### How big is DeepSeek-R1-Zero and what license is it under?

It is a Mixture-of-Experts model with 671B total parameters and 37B activated per token, supporting a 128K-token context window and up to 32,768 output tokens. It is released under the permissive MIT license, allowing commercial use and derivative works.

## Links

- [DeepSeek-R1-Zero on Hugging Face](https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero)
- [DeepSeek-R1 paper (arXiv 2501.12948)](https://arxiv.org/html/2501.12948v1)
- [DeepSeek-R1 GitHub repository](https://github.com/deepseek-ai/DeepSeek-R1)
- [Nature: DeepSeek-R1 peer-reviewed publication](https://www.nature.com/articles/s41586-025-09422-z)
- [DeepSeek — official site](https://www.deepseek.com)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/deepseek-r1-zero/