How to run DeepSeek AI locally to protect your privacy - 2 easy ways

3. Install DeepSeek
4. Select and Use DeepSeek
How to Install DeepSeek Locally from the Linux Command Line
Alternatively, you can perform a full installation of DeepSeek on a Linux system. Be aware that this method has some hefty system requirements:
- CPU: A powerful multi-core processor with at least 12 cores recommended.
- GPU: An NVIDIA GPU with CUDA support for better performance. If Ollama doesn't detect an NVIDIA GPU, it will switch to CPU-only mode.
- RAM: At least 16 GB, with 32 GB or more preferred.
- Storage: NVMe storage for faster operations.
- Operating System: Ubuntu or an Ubuntu-based distribution.
If your system meets these requirements and you have Ollama installed, you can run the DeepSeek R1 model with the command:
ollama run deepseek-r1:8b
If you haven't installed Ollama yet, you can do so with:
curl -fsSL https://ollama.com/install.sh | sh
You'll be asked for your user password during the installation.
There are several versions of DeepSeek you can run, including:
ollama run deepseek-r1 - The default 8B versionollama run deepseek-r1:1.5b - The smallest modelollama run deepseek-r1:7b - The 7B versionollama run deepseek-r1:14b - The 14B versionollama run deepseek-r1:32b - The 32B versionollama run deepseek-r1:70b - The largest and most advanced model
Once the command finishes, you'll be at the Ollama prompt, ready to use your chosen model.
With either method, you'll have access to DeepSeek AI, allowing you to use it securely on your local machine, keeping your queries and data private.
Related article
AI Search Mandatory Policy Fuels Exodus, DuckDuckGo Sees User Surge
Following Google's 2026 I/O conference announcement of a full AI overhaul of its search engine, many users started looking for more controllable alternatives because there was no simple "one-click disable" for AI features. The privacy-focused search
Xiaohongshu Restructures: Conan Named President, Creates AI Primary Department Dots and Overseas Division Rednote
On April 30, Xiaohongshu sent an internal memo to all employees announcing the launch of a new organizational restructuring. The core of this change involves fully integrating three business lines—community, e-commerce, and commercialization—along wi
Tencent's Xiaolongxia Surges Beyond Expectations, Team Expands Capacity 10x, Apologizes and Compensates
Tencent has officially launched WorkBuddy, an all-scenario AI intelligent agent, marking a new phase in the large model application layer race with high integration and a low deployment threshold.The product drew immediate industry attention on its l
Related Special Topic Recommendations
Comments (15)
0/500
로컬에서 AI를 실행하는 방법에 대한 글 정말 도움 됐어요! 개인정보 보호 이슈가 점점 중요해지는 시대에 DeepSeek 같은 오픈소스 모델이 큰 대안이 될 것 같네요. 한국에서도 AI 개발 경쟁이 치열한데, 비용 효율적인 중국 모델이 시장에 어떤 영향을 미칠지 궁금해요 🤔 혹시 한국어 지원은 언제쯤 될까요?
Also das mit der lokalen Installation ist ja wirklich praktisch! Endlich muss man nicht mehr ständig Daten in die Cloud schicken. Aber mal ehrlich, wie viel Speicherplatz braucht so ein Modell eigentlich? Mein Laptop ist schon jetzt voll mit Fotos... 🤔
¡Qué interesante! No sabía que DeepSeek era una startup china con costes tan bajos. Me pregunto cómo logran desarrollar modelos tan efectivos con menos recursos que sus competidores estadounidenses. La opción de ejecución local es crucial para la privacidad de datos, especialmente para aquellos que trabajan con información sensible. ¿Alguien ya lo ha probado? Me encantaría conocer experiencias reales 🧐
DeepSeek running locally sounds cool, but I'm skeptical about the setup complexity. Has anyone tried it yet? 🤔
Super cool to see DeepSeek making waves with affordable LLMs! 😎 Running it locally sounds like a privacy win, but I’m curious—how does it stack up against giants like GPT? Anyone tried it yet?

3. Install DeepSeek
4. Select and Use DeepSeek
How to Install DeepSeek Locally from the Linux Command Line
Alternatively, you can perform a full installation of DeepSeek on a Linux system. Be aware that this method has some hefty system requirements:
- CPU: A powerful multi-core processor with at least 12 cores recommended.
- GPU: An NVIDIA GPU with CUDA support for better performance. If Ollama doesn't detect an NVIDIA GPU, it will switch to CPU-only mode.
- RAM: At least 16 GB, with 32 GB or more preferred.
- Storage: NVMe storage for faster operations.
- Operating System: Ubuntu or an Ubuntu-based distribution.
If your system meets these requirements and you have Ollama installed, you can run the DeepSeek R1 model with the command:
ollama run deepseek-r1:8bIf you haven't installed Ollama yet, you can do so with:
curl -fsSL https://ollama.com/install.sh | shYou'll be asked for your user password during the installation.
There are several versions of DeepSeek you can run, including:
ollama run deepseek-r1- The default 8B versionollama run deepseek-r1:1.5b- The smallest modelollama run deepseek-r1:7b- The 7B versionollama run deepseek-r1:14b- The 14B versionollama run deepseek-r1:32b- The 32B versionollama run deepseek-r1:70b- The largest and most advanced model
Once the command finishes, you'll be at the Ollama prompt, ready to use your chosen model.
With either method, you'll have access to DeepSeek AI, allowing you to use it securely on your local machine, keeping your queries and data private.
AI Search Mandatory Policy Fuels Exodus, DuckDuckGo Sees User Surge
Following Google's 2026 I/O conference announcement of a full AI overhaul of its search engine, many users started looking for more controllable alternatives because there was no simple "one-click disable" for AI features. The privacy-focused search
Xiaohongshu Restructures: Conan Named President, Creates AI Primary Department Dots and Overseas Division Rednote
On April 30, Xiaohongshu sent an internal memo to all employees announcing the launch of a new organizational restructuring. The core of this change involves fully integrating three business lines—community, e-commerce, and commercialization—along wi
Tencent's Xiaolongxia Surges Beyond Expectations, Team Expands Capacity 10x, Apologizes and Compensates
Tencent has officially launched WorkBuddy, an all-scenario AI intelligent agent, marking a new phase in the large model application layer race with high integration and a low deployment threshold.The product drew immediate industry attention on its l
로컬에서 AI를 실행하는 방법에 대한 글 정말 도움 됐어요! 개인정보 보호 이슈가 점점 중요해지는 시대에 DeepSeek 같은 오픈소스 모델이 큰 대안이 될 것 같네요. 한국에서도 AI 개발 경쟁이 치열한데, 비용 효율적인 중국 모델이 시장에 어떤 영향을 미칠지 궁금해요 🤔 혹시 한국어 지원은 언제쯤 될까요?
Also das mit der lokalen Installation ist ja wirklich praktisch! Endlich muss man nicht mehr ständig Daten in die Cloud schicken. Aber mal ehrlich, wie viel Speicherplatz braucht so ein Modell eigentlich? Mein Laptop ist schon jetzt voll mit Fotos... 🤔
¡Qué interesante! No sabía que DeepSeek era una startup china con costes tan bajos. Me pregunto cómo logran desarrollar modelos tan efectivos con menos recursos que sus competidores estadounidenses. La opción de ejecución local es crucial para la privacidad de datos, especialmente para aquellos que trabajan con información sensible. ¿Alguien ya lo ha probado? Me encantaría conocer experiencias reales 🧐
DeepSeek running locally sounds cool, but I'm skeptical about the setup complexity. Has anyone tried it yet? 🤔
Super cool to see DeepSeek making waves with affordable LLMs! 😎 Running it locally sounds like a privacy win, but I’m curious—how does it stack up against giants like GPT? Anyone tried it yet?





Home






