DEV Community

0 seconds of 0 secondsVolume 90%
Press shift question mark to access a list of keyboard shortcuts
00:00
00:00
00:00
 
BHUVANESH M
BHUVANESH M

Posted on

2

🧠 Run DeepSeek Offline on Ubuntu (CPU/GPU) Using Ollama – Quick Guide

If you're looking to run DeepSeek (specifically deepseek-coder) offline on your Ubuntu machine without worrying about GPUs, you're in the right place!

✅ Works even without a GPU
✅ Just 2 simple commands
✅ Completely offline usage after download


🔧 Step 1: Install Ollama

Run this command in your terminal to install Ollama:

curl -fsSl https://ollama.com/install.sh | sh
Enter fullscreen mode Exit fullscreen mode

✅ This installs Ollama and sets it up as a system service.

📌 You may see a message like:

WARNING: No NVIDIA/AMD GPU detected. Ollama will run in CPU-only mode.
Enter fullscreen mode Exit fullscreen mode

No problem! It still works perfectly on CPU-only systems!


🧐 Step 2: Run DeepSeek Coder Locally

Now, run the DeepSeek LLM (offline) using:

ollama run deepseek-coder
Enter fullscreen mode Exit fullscreen mode

Ollama will automatically download the model the first time and run it locally after that.


💻 Features

  • Runs entirely offline after the first model download
  • Works on CPU or GPU
  • Lightweight installation
  • Perfect for offline development, coding, or learning AI locally

🎮 YouTube Tutorial

If the embed does not render correctly, you can also watch it here:

👉 Watch on YouTube


💬 Questions?

Feel free to comment or connect with me on LinkedIn or follow more updates on bhuvaneshm.in!

DevCycle image

Ship Faster, Stay Flexible.

DevCycle is the first feature flag platform with OpenFeature built-in to every open source SDK, designed to help developers ship faster while avoiding vendor-lock in.

Start shipping

Top comments (1)

Collapse
 
bhuvaneshm_dev profile image
BHUVANESH M

Now that DeepSeek is running offline, what other LLMs should we try next on Ollama? Maybe code-specific models like Code Llama, or try fine-tuning a small model locally? Drop your ideas below!

Redis image

Short-term memory for faster
AI agents

AI agents struggle with latency and context switching. Redis fixes it with a fast, in-memory layer for short-term context—plus native support for vectors and semi-structured data to keep real-time workflows on track.

Start building

👋 Kindness is contagious

Explore this compelling article, highly praised by the collaborative DEV Community. All developers, whether just starting out or already experienced, are invited to share insights and grow our collective expertise.

A quick “thank you” can lift someone’s spirits—drop your kudos in the comments!

On DEV, sharing experiences sparks innovation and strengthens our connections. If this post resonated with you, a brief note of appreciation goes a long way.

Get Started