nv-generate-ct-rflow
nvidiaGenerate synthetic CT volumes and paired masks.
medical imaging
What it does
Generates synthetic CT volumes and masks through the NV-Generate-CTMR rflow-ct wrapper.
When to use it
Use when synthetic CT volumes and masks are needed for reviewed, non-production research.
How to use it
Give it a config_infer override and an output directory; it writes synthetic CT volumes, masks, and result JSON there.
What you provide
- a config_infer override file
- an output directory
- Your files
Uses
Access · 8
This skill
https://huggingface.co
Downloads setup assets from Hugging Face
https://github.com
Downloads setup files from GitHub
NVIDIA-Medtech/NV-Generate-CTMR
Installs a pinned upstream checkout
$NV_GENERATE_ROOT/requirements.txt
Installs upstream Python packages
What you need · 4
Python must run the package installation and wrapper commands.
rflow-ct requires CUDA and has no CPU fallback.
Requires an NVIDIA GPU with at least 16 GB VRAM; larger outputs need an A100 or H100.
Requires a local checkout containing the rflow-ct weights and mask-candidate datasets.
About this skill
- Visibility
- Public
- Repository
- nvidia/skills
- Created
- Oct 8, 2026
- Updated
- Oct 8, 2026
- Files
- 33