Transparent Pricing

Unlimited labeling.
Zero cloud limits.

Cloud annotation platforms charge per image, per credit, or per team seat. SentinelLabeler runs 100% on your hardware with zero artificial caps.

Monthly Plan
Pay as you go. Perfect for short-term dataset sprints or validating model concepts.
$15/ month
  • Unlimited dataset sizes & video frames
  • Segment Anything 2.1 (SAM 2.1) auto-masking
  • YOLO-World zero-shot text prompting
  • Video tools: tracking (up to 60 frames), copy previous, fill between, label similar
  • Dual-model live A/B compare
  • SAHI high-resolution tiled inference
  • Local NVIDIA CUDA GPU training studio
  • Train YOLO11, YOLO26 & RT-DETR; export models as ONNX
  • 1-click hot-swap into Sentinel Core Vision
  • Cancel anytime with 1 click in your account
Try Monthly free for 7 days →

7-day free trial for new customers. Card required; cancel before day 7 and you are not charged.

Secure Stripe checkout Cancel anytime with 1 click Runs 100% on your machine Models & datasets remain yours forever
Frequently Asked Questions

Everything you need to know

What happens if I cancel my subscription?

You keep all of your datasets, labels, and trained models forever. Because SentinelLabeler stores everything in standard formats (YOLO text files, Pascal VOC XML, COCO JSON, PyTorch .pt, and ONNX) locally on your hard drive, you will never be locked out of your research or data. If you cancel, your license remains active until the end of your billing cycle.

Can I use my license on more than one computer?

Yes. Each license key includes device activation rights. If you build a new PC or switch between your capture workstation and your training rig, you can easily release and reactivate your license from your account dashboard without waiting for manual support.

Does SentinelLabeler upload any of my video footage or datasets?

Never. SentinelLabeler is fundamentally zero-cloud. All AI inference (SAM 2.1, YOLO-World, SAHI) and all model training runs directly on your local GPU via your isolated PyTorch environment. The only network calls made by SentinelLabeler are periodic license status checks to our verification endpoint.

Do I need an NVIDIA GPU to run SentinelLabeler?

For manual bounding-box annotation and video navigation, an ordinary modern multi-core CPU is sufficient. However, for AI auto-annotation (SAM 2.1) and Local Training Studio, an NVIDIA RTX GPU with at least 6GB of VRAM (RTX 2060 or higher recommended) is required for real-time acceleration via CUDA 12.x.

How does the free trial work?

New customers get the first 7 days free on the monthly or yearly plan. You add a card at checkout, but nothing is charged during the trial. On day 7 your plan starts and the card is charged, unless you cancel before then: Account, then Manage billing. Cancel during the trial and you pay nothing.

The free trial is once per customer: one per email, one per payment card and one per PC. If you had SentinelLabeler before, checkout starts your plan straight away.

How does the refund policy work?

We stand behind SentinelLabeler with a clear, honest refund policy. If the software fails to operate on your hardware due to an unresolved technical incompatibility within 7 days of your initial purchase, our support squad will either assist you in configuring it or process a full refund. See our Refund Policy for complete details.

Ready to accelerate your labeling?

Join computer vision engineers and creators producing state-of-the-art YOLO models on their own hardware.