Back to feed
Dev.to
Dev.to
7/11/2026
Setting Up a Local AI Coding Agent with Ollama and Aider

Setting Up a Local AI Coding Agent with Ollama and Aider

Short summary

A practical guide to running a fully local AI coding agent using Ollama on Windows and Aider inside WSL2 Ubuntu. The author details a CPU-only setup with qwen2.5-coder:14b, including WSL mirrored networking configuration to connect Aider to Ollama. The architecture keeps models on Windows while repos live in the Linux filesystem, avoiding duplicate installations and maintaining privacy.

  • •Ollama runs on Windows serving qwen2.5-coder:14b; Aider runs in WSL2 Ubuntu connecting via localhost:11434
  • •WSL mirrored networking mode required for Aider to reach Ollama on Windows
  • •CPU-only inference works but is slower; setup validated on i7 with 48GB RAM and no dedicated GPU

Generated with AI, which can make mistakes.

#ai-tools#ai-agents#open-source
Read full article at Dev.to

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

Build local AI coding agents
Dev.to
Running Local AI Models in .NET with Ollama (Step-by-Step Guide)
Dev.to
The original headline is: "Run Qwen Coder & DeepSeek Locally: The 2026 Free AI Pair-Programmer Setup"
Dev.to
Build Your Own Local AI Coding Agent with Gemma 4 and OpenCode
Towards Data Science