T-Rex Label
Verified Oct 15, 2026 ActiveFree browser-based annotation tool that labels objects across a dataset from one example.
Listed on EffectHub ·
Pricing
- Pricing model
- 100% Free
Listed as free to use; no paid tier or price was found published on the site.
Use tool ↗About T-Rex Label
T-Rex Label is a browser-based annotation tool for building computer-vision datasets, using a zero-shot detection model so it can identify and label objects without training or fine-tuning first. A user draws one bounding box or segmentation mask, and cross-image inference applies that same visual prompt across the rest of the dataset to pre-annotate similar objects automatically. It imports and exports common annotation formats and connects to Kaggle Datasets, Hugging Face, Label Studio, PyTorch and TensorFlow. The tool requires no installation and is offered free of charge, built by Vision Qimeng (Shenzhen) Technology Co., Ltd. in China as a companion product to its T-Rex2 object-detection model.
- Platforms
- Web
- API
- Not confirmed
- Last verified
- Oct 15, 2026
Verified sources
- Product features, integrations and free pricingtrexlabel.com ↗
- Free-of-charge confirmation and use cases from the model's own teamdeepdataspace.com ↗
Is T-Rex Label legit?
We don’t label tools safe or unsafe. Below are the records we checked ourselves, each with the date we read it.
- Domain registered
- May 9, 2024 (2 years ago)Registry record (RDAP), registrar Alibaba Cloud Computing Ltd. d/b/a HiChina (www.net.cn), looked up Sep 6, 2026
- TLS certificate
- Issued by Let's Encrypt, valid until Nov 24, 2026Seen in our own connection to the site on Oct 15, 2026
- Last checked
- Oct 15, 2026The site answered
- Operated by
- Vision Qimeng (Shenzhen) Technology Co., Ltd., ChinaOur research, card verified Sep 1, 2026
Pros & cons of T-Rex Label
T-Rex Label's cross-image inference from a single labeled example can meaningfully cut the time needed to build a computer-vision dataset, and offering it free removes the usual cost barrier to trying a new annotation tool. It works best on datasets with visually consistent, repeated objects; more varied scenes will still need manual correction.
Pros
- Zero-shot detection skips the model training step most annotation tools require first
- Cross-image inference propagates one labeled example across an entire dataset
- No installation required and free to use
Cons
- Pre-annotation accuracy will vary with how visually consistent the target objects are
- No published enterprise pricing or support tier for teams needing guarantees
- Company is not widely known outside computer-vision circles, so long-term support is unproven
Best for
- Computer vision teams building datasets with repeated, visually similar objects
- Teams already using Label Studio, Hugging Face or Kaggle who want a compatible free annotator
Not for
- Datasets with highly varied, dissimilar objects where one-shot prompting won't generalize
- Teams needing enterprise support contracts or guaranteed SLAs, which aren't advertised
Alternatives to T-Rex Label
T-Rex Label head-to-head
Side-by-side fact sheets against the competitors our research or their makers name — pricing, platforms, API and both verdicts on one page.