deepseek-r1:7b - Ollama 框架

deepseek-r1:7b

78.2M 下载量更新于 7 个月前

DeepSeek-R1 是一系列具有领先模型性能的开源推理模型，例如 O3 和 Gemini 2.5 Pro。

工具思考 1.5b 7b 8b 14b 32b 70b 671b

ollama run deepseek-r1:7b

curl https://:11434/api/chat \
  -d '{
    "model": "deepseek-r1:7b",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='deepseek-r1:7b',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'deepseek-r1:7b',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

详情

8个月前更新

8个月前

755ced02ce7b · 4.7GB ·

模型

架构qwen2

·

参数7.62B

·

量化Q4_K_M

4.7GB

许可

MIT 许可证版权所有 (c) 2023 DeepSeek 经此特此授予任何人免费使用许可

1.1kB

参数

{ "stop": [ "<｜begin▁of▁sentence｜>", "<｜end▁of▁sentence｜>",

148B

模板

{{- if .System }}{{ .System }}{{ end }} {{- range $i, $_ := .Messages }} {{- $last := eq (len (slice

556B

自述文件

DeepSeek-R1 已经收到小版本升级至 DeepSeek-R1-0528，适用于 80 亿参数蒸馏模型和完整的 6710 亿参数模型。在此更新中，DeepSeek R1 在推理和推断能力方面有了显著提升。该模型在各种基准评估中表现出色，包括数学、编程和通用逻辑。其整体性能现在接近领先模型，例如 O3 和 Gemini 2.5 Pro。

模型

DeepSeek-R1-0528-Qwen3-8B

ollama run deepseek-r1

DeepSeek-R1

ollama run deepseek-r1:671b

注意：要从旧版本更新模型，请运行 ollama pull deepseek-r1

蒸馏模型

DeepSeek 团队已经证明，较大模型的推理模式可以蒸馏到较小模型中，从而获得比通过在小型模型上进行强化学习发现的推理模式更好的性能。

以下是使用 DeepSeek-R1 生成的推理数据，通过对研究界广泛使用的几种密集模型进行微调而创建的模型。评估结果表明，蒸馏后的较小密集模型在基准测试中表现出色。

DeepSeek-R1-0528-Qwen3-8B

ollama run deepseek-r1:8b

DeepSeek-R1-Distill-Qwen-1.5B

ollama run deepseek-r1:1.5b

DeepSeek-R1-Distill-Qwen-7B

ollama run deepseek-r1:7b

DeepSeek-R1-Distill-Qwen-14B

ollama run deepseek-r1:14b

DeepSeek-R1-Distill-Qwen-32B

ollama run deepseek-r1:32b

DeepSeek-R1-Distill-Llama-70B

ollama run deepseek-r1:70b

许可协议

模型权重在 MIT 许可证下授权。DeepSeek-R1 系列支持商业用途，允许进行任何修改和衍生作品，包括但不限于用于训练其他 LLM 的蒸馏。请注意，

Qwen 蒸馏模型源自 Qwen-2.5 系列，最初在 Apache 2.0 许可证下授权，现在使用 DeepSeek-R1 策划的 80 万个样本进行微调。

Llama 8B 蒸馏模型源自 Llama3.1-8B-Base，最初在 llama3.1 许可证下授权。

Llama 70B 蒸馏模型源自 Llama3.3-70B-Instruct，最初在 llama3.3 许可证下授权。

<img src="/assets/library/deepseek-v3/069ccc94-63b0-41e6-b2b3-e8e56068ab1a" width="320" />

DeepSeek-R1 has received a minor version upgrade to DeepSeek-R1-0528 for the 8 billion parameter distilled model and the full 671 billion parameter model. In this update, DeepSeek R1 has significantly improved its reasoning and inference capabilities. The model has demonstrated outstanding performance across various benchmark evaluations, including mathematics, programming, and general logic. Its overall performance is now approaching that of leading models, such as O3 and Gemini 2.5 Pro.

![image.png](/assets/library/deepseek-r1/ba9ac535-ac57-4d42-9f36-20067c8eaa50)

## Models

**DeepSeek-R1-0528-Qwen3-8B**

```
ollama run deepseek-r1
```

**DeepSeek-R1**

```
ollama run deepseek-r1:671b
```

> Note: to update the model from an older version, run `ollama pull deepseek-r1`

### Distilled models

DeepSeek team has demonstrated that the reasoning patterns of larger models can be distilled into smaller models, resulting in better performance compared to the reasoning patterns discovered through RL on small models.

Below are the models created via fine-tuning against several dense models widely used in the research community using reasoning data generated by DeepSeek-R1. The evaluation results demonstrate that the distilled smaller dense models perform exceptionally well on benchmarks.

**DeepSeek-R1-0528-Qwen3-8B**

```
ollama run deepseek-r1:8b
```

**DeepSeek-R1-Distill-Qwen-1.5B**

```
ollama run deepseek-r1:1.5b
```

**DeepSeek-R1-Distill-Qwen-7B**

```
ollama run deepseek-r1:7b
```

**DeepSeek-R1-Distill-Qwen-14B**

```
ollama run deepseek-r1:14b
```

**DeepSeek-R1-Distill-Qwen-32B**

```
ollama run deepseek-r1:32b
```

**DeepSeek-R1-Distill-Llama-70B**

```
ollama run deepseek-r1:70b
```

### License

The model weights are licensed under the MIT License. DeepSeek-R1 series support commercial use, allow for any modifications and derivative works, including, but not limited to, distillation for training other LLMs. Please note that:

The Qwen distilled models are derived from Qwen-2.5 series, which are originally licensed under Apache 2.0 License, and now finetuned with 800k samples curated with DeepSeek-R1.

The Llama 8B distilled model is derived from Llama3.1-8B-Base and is originally licensed under llama3.1 license.

The Llama 70B distilled model is derived from Llama3.3-70B-Instruct and is originally licensed under llama3.3 license.

粘贴、拖放或单击以上传图像（.png、.jpeg、.jpg、.svg、.gif）