How to Install Miniconda on Windows
Install Miniconda on Windows if you want conda environments, isolated Python versions, and easier package installs for data science or native dependencies.
Supported Versions
Official Source
anaconda.com (official)
Environment Setup
3 PATH entries
Verify Step
conda --version
DevTools Installer is not affiliated with, endorsed by, or sponsored by the publisher of Miniconda. All product names and logos are trademarks of their respective owners. Downloads are sourced directly from the official publisher.
Install With DevTools Installer
Use this path if you want DevTools Installer to download the official package and handle the Windows setup for you.
- 1Open DevTools Installer.
- 2In Package Managers, select "Miniconda".
- 3Click "Install Selected".
- 4Let the installer download the official setup package and run an all-users install to .
- 5If an earlier broken install is detected, DevTools Installer removes it before retrying the install.
- 6After the install finishes, the root, Scripts, and condabin directories are added to PATH.
- 7Open a new terminal and run .
Download Source
DevTools Installer downloads from the official publisher:
https://repo.anaconda.com/miniconda/Miniconda3-latest-Windows-x86_64.exe
PATH & Environment Variables
DevTools Installer sets the following automatically:
PATH entries
C:\ProgramData\miniconda3
C:\ProgramData\miniconda3\Scripts
C:\ProgramData\miniconda3\condabin
Verification
DevTools Installer verifies the install by running:
conda --version
Expected output: conda 26.x.x
Manual Installation Steps
Use these steps if you want to run the installer yourself and apply the Windows PATH or environment changes manually.
- 1Download the official installer from https://repo.anaconda.com/miniconda/Miniconda3-latest-Windows-x86_64.exe
- 2Run the installer and choose "Install for all users" if you want the standard path .
- 3If you need a silent install, run:
- 4Add these three directories to PATH: , , and .
- 5Open a new terminal and run .
- 6If the installer fails because an old broken folder already exists, delete and retry.
Need to upgrade, downgrade, or remove Miniconda?
DevTools Installer can also upgrade Miniconda to a newer version, roll back to an older one, or cleanly uninstall it — including PATH entries and environment variables. No manual cleanup required.
What Is Miniconda Used For?
- Setting up isolated Python environments for data science, machine learning, and scientific computing.
- Installing packages that have complex native dependencies (NumPy, SciPy, TensorFlow) — conda handles binary dependencies that pip cannot.
- Managing multiple Python versions side by side with "conda create -n myenv python=3.12".
- Sharing reproducible environments across teams using environment.yml files.
- Working with Jupyter notebooks and JupyterLab in isolated conda environments.
Common Issues And Fixes
Check the problem and the exact fix before you reinstall anything. Some guides also include the reason the issue happens.
Problem
"conda" is not recognized after installation
Exact fix
Ensure all three directories are in PATH: C:\ProgramData\miniconda3, its Scripts subfolder, and its condabin subfolder. DevTools Installer adds these automatically, but manual installs may miss them.
Problem
Conda activate does not work in Command Prompt
Exact fix
Run "conda init cmd.exe" once to set up Command Prompt integration. For PowerShell, run "conda init powershell". Restart the terminal after initialization.
Problem
Installation fails because a previous corrupt installation exists
Exact fix
Delete the C:\ProgramData\miniconda3 directory manually, then retry the installation. DevTools Installer detects and cleans up corrupt installs automatically.
Problem
Conda is very slow to resolve dependencies
Exact fix
Use the libmamba solver (default in conda 23.10+) for faster resolution: "conda config --set solver libmamba". Also consider creating minimal environments instead of installing everything into base.
Tips for Miniconda
- ✓Miniconda is much smaller than the full Anaconda distribution — it includes only conda, Python, and a minimal set of packages. Install additional packages as needed.
- ✓If you already have a standalone Python installed, Miniconda creates separate Python installations per environment. The two can coexist without conflict.
- ✓Use "conda env export > environment.yml" to save your environment configuration, and "conda env create -f environment.yml" to recreate it on another machine.
- ✓For data science workflows, install JupyterLab inside a conda environment: "conda install jupyterlab".
Related Guides
Python
Install Python on Windows so you can run scripts, create virtual environments, and use pip without the Microsoft Store redirect getting in the way.
Visual Studio Code
Install VS Code on Windows if you want a fast editor with extensions, terminal integration, and the code command available on PATH.
Git
Install Git on Windows so you can clone repositories, commit changes, use Git Bash, and run git from any terminal.