Documentation
MultiCellPose
Installation

Installation

License and Attribution Notice
MultiCellPose includes and modifies components from Cellpose.
Original copyright © 2020 Howard Hughes Medical Institute.
Original contributors include Marius Pachitariu and Carsen Stringer.
Distributed under the BSD 3-Clause License.
This project is an independent derivative work and is not endorsed by HHMI or the original contributors.
Full license text: MultiCellPose License

1. Installation Recommended (PyPI)

Step 1: Open Anaconda Prompt (Windows) or terminal (Linux/macOS)

Create and activate a clean environment:

conda create -n multicellpose python=3.10 -y
conda activate multicellpose

If Conda prompts for Terms of Service acceptance, run:

conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/main
conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/r
conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/msys2

Step 2: Install PyTorch first (choose CPU or GPU)

GPU users: check your CUDA version first (Windows):

nvidia-smi

Look for CUDA Version: X.Y. Then install the matching PyTorch wheel:

  • CUDA 12.6+:
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126
  • CUDA 12.1:
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
  • CUDA 11.8:
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

CPU-only:

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu

Step 3: Install MultiCellPose 

pip install multicellpose[gui]

This installs GUI, server/client communication dependencies, and image readers (.nd2, .lif, .nrrd) by default.

Step 4: Launch 

multicellpose-gui

Manual installation (if you will edit the code)

If Git is not installed, download it from:

Open Git CMD and clone the repo

  1. Open Git CMD (Start Menu -> search “Git CMD”).
  2. Choose a folder where you want the project to live (example: C:\Projects).
  3. In Git CMD, run:
cd C:\Projects
git clone https://github.com/mrcsfltchr/multicellposerelease.git
cd MultiCellPose

2. Install Miniconda and create an environment

Download Miniconda (Windows x86_64):

Open Anaconda Prompt

  1. Open Anaconda Prompt (Start Menu -> search “Anaconda Prompt”).
  2. Navigate to the repo folder (example below):
cd C:\Projects\MultiCellPose

Create and activate the environment:

conda create -n multicellpose python=3.10 -y
conda activate multicellpose

If Conda prompts for Terms of Service acceptance, run:

conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/main
conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/r
conda tos accept --override-channels --channel https://repo.anaconda.com/pkgs/msys2

3. Install PyTorch (CPU or GPU)

Check your CUDA version:

nvidia-smi

Install the matching PyTorch wheel:

# CUDA 12.6+
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu126

# CUDA 12.1
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121

# CUDA 11.8
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

# CPU-only
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu

4. Install dependencies and the app 

pip install numpy scipy pandas scikit-image pyqt6 pyqt6.sip qtpy pyqtgraph superqt natsort tifffile fastremap tqdm matplotlib
pip install grpcio grpcio-tools protobuf
pip install -e ".[all]"

5. Launch the GUI 

python -m guv_app.main