# DeepSeek R1

DeepSeek R1 is an AI model released by DeepSeek on Jan 20 2025. It has 671B parameters, a 128k context window and open weight. Tracked results include 71.5% on GPQA Diamond, 49.2% on SWE-Bench Verified and 65.9% on LiveCodeBench.

## Facts

| Field | Value |
| --- | --- |
| Model | DeepSeek R1 |
| Developer | DeepSeek |
| Release date | Monday, Jan 20 2025 |
| Licensing | Open Weight |
| Parameters | 671B |
| Context window | 128k |

## Tracked benchmark scores

| Benchmark | Score | Source | What it measures |
| --- | --- | --- | --- |
| SWE-Bench Verified | 49.2% | Lab | Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. |
| LiveCodeBench | 65.9% | Lab | Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. |
| MATH-500 | 97.3% | Lab | accuracy on the 500-problem MATH subset |
| DROP | 92.2 | Lab | three-shot token F1 on questions requiring reasoning over paragraphs |
| AIME 2024 | 79.8% | Lab | single-answer accuracy on the 2024 invitational mathematics exam without tools |
| GPQA Diamond | 71.5% | Lab | Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. |
| SimpleQA | 30.1% | Lab | accuracy on short, fact-seeking questions in the original SimpleQA set |
| MMLU-Pro | 84% | Lab | accuracy on the ten-option MMLU-Pro benchmark |
| MMLU | 90.8% | Lab | A 57-subject multiple-choice exam — history, law, medicine, maths — that was the standard measure of how much a model knows from 2020 until roughly 2024, when frontier scores crowded into the high 80s and labs moved on to harder tests. The scores here were published years apart under different testing setups, so read them as a historical record rather than a like-for-like ranking. Higher is better. |

## About DeepSeek R1

DeepSeek R1 was released in January 2025 as a reasoning model built on DeepSeek V3 Base. DeepSeek published the main model weights under the MIT License and described a training process combining cold-start supervised data with reinforcement learning. Its research also introduced R1-Zero, an experiment that applied reinforcement learning without preliminary supervised fine-tuning.

R1 and R1-Zero were distinct checkpoints: the final R1 recipe addressed readability and language-mixing problems observed in R1-Zero. The release also included six smaller distilled models based on Qwen and Llama foundations. Those variants made the reasoning work available at different model sizes, with the underlying foundation-model licenses relevant to their use.

## Questions and answers

### When was DeepSeek R1 released?

DeepSeek R1 was released by DeepSeek on Monday, Jan 20 2025.

### Who made DeepSeek R1?

DeepSeek R1 was built by DeepSeek. Chinese AI lab known for efficient, open-weight models. Gained attention for strong performance at lower cost.

### What benchmark scores did DeepSeek R1 get?

DeepSeek R1 reports 9 tracked benchmark scores — SWE-Bench Verified: 49.2%; LiveCodeBench: 65.9%; MATH-500: 97.3%; DROP: 92.2; AIME 2024: 79.8%; GPQA Diamond: 71.5%; SimpleQA: 30.1%; MMLU-Pro: 84%; MMLU: 90.8%. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data.

### What is the context window of DeepSeek R1?

DeepSeek R1 has a context window of 128k. Providers may specify separate input and output token limits.

### How many parameters does DeepSeek R1 have?

DeepSeek R1 is reported at 671B parameters.

### Is DeepSeek R1 open source?

Partly. DeepSeek R1 is an open-weight model: the trained weights are free to download and run locally or on your own infrastructure, but the training data and code are not fully released and the license may restrict some commercial uses. It is not open source in the strict sense.

### What came before and after DeepSeek R1?

DeepSeek's previous tracked release was DeepSeek V3 on Dec 26 2024, 25 days earlier. It was followed by DeepSeek R1 Distill on Jan 20 2025.


---

Canonical page: https://aireleasetracker.com/model/deepseek/deepseek-r1
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.
