Quick Start

Getting Started

Install QCoder, configure your first AI provider, and start coding with AI assistance in minutes.

System Requirements

Before installing QCoder, make sure your system meets the following requirements:

Requirement	Minimum	Recommended
---	---	---
Operating System	Windows 10 64-bit	Windows 11 64-bit
RAM	8 GB	16 GB or more
Disk Space	500 MB	1 GB (including models cache)
Node.js	18.x or later	22.x LTS
Git	2.x	Latest stable
Docker	Optional	Recommended for sandbox mode

QCoder runs as a standalone desktop application based on VS Code. If you can run VS Code, you can run QCoder.

Installation

QCoder is distributed as source code. Clone the repository from [GitHub](https://github.com/Quantum-Horizon-AI/QCoder) and build it with the steps in the [README](https://github.com/Quantum-Horizon-AI/QCoder#build-from-source). The build process is the same as VS Code's, with a few QCoder patches.

QCoder is free and open source under the MIT License. There is no license key and no account.

First Launch

When you open QCoder for the first time, you will see the standard editor layout with a sidebar on the left. The AI chat panel lives in the sidebar -- look for the QCoder icon.

Step-by-step setup:

Open the Chat Panel -- Click the QCoder icon in the activity bar (sidebar), or press Ctrl+Shift+P and type QCoder: Focus Chat.
Open Settings -- Click the gear icon at the top of the chat panel. This opens the Settings view.
Go to API Config -- Select the API Config tab at the top of Settings.
Select a Provider -- Choose your AI provider from the dropdown (e.g., OpenAI, Anthropic, Google Gemini, DeepSeek, Ollama).
Enter Your API Key -- Paste your API key into the field. Keys are stored locally in your OS credential store and are never sent to QCoder servers.
Choose a Model -- The model dropdown populates automatically based on your provider. Select a model (e.g., gpt-4o, claude-sonnet-4-20250514, gemini-2.0-flash).
Start Chatting -- Go back to the Chat tab, type a message, and press Enter. You are ready to code with AI.

If you are using a local model runner like Ollama or LM Studio, no API key is needed. Just point QCoder at the local endpoint (usually http://localhost:11434 for Ollama).

Opening a Project

QCoder works best when you have a workspace folder open:

File > Open Folder -- Select your project root directory.
Drag and drop -- Drag a folder from your file manager onto the QCoder window.
Command line -- Run qcoder /path/to/project from a terminal.

Once a workspace is open, QCoder's AI tools have access to your file tree, enabling features like file reading, code search, and workspace-aware suggestions.