This website uses cookies We use cookies to personalise content and ads, to provide social media features and to analyse our traffic. We also share information about your use of our site with our social media, advertising and analytics partners who may combine it with other information that you’ve provided to them or that they’ve collected from your use of their services. See our cookie policy.

We make Machine Learning
Reproducible and Safe

You choose a model and we take care of the rest:
running at scale and benchmarking ML models.

Modal video thumbnail

Built for and trusted by ML engineers, researchers and academics.


Backed By


We accelerate your AI journey

We are making a shortcut for ML teams everywhere.
Speed up your development cycle: Run, test and improve your ML models with just a few clicks.

Focus on building ML models

We accelerate the development cycle from prototyping to deployment by dealing with the hassle of running your models at scale and benchmarking them against other models and data sets.

AI safety

Significantly improve model performance by uncovering and fixing model failures.


Know how well your models perform compared to other models & data sets.

Optimize Cloud Spend

We efficiently deploy your models - only pay what you use.

Evaluate across many fields

Benchmarking is available for a selected group of categories for now.

Object Detection

PASCAL VOC 2007 - [link]COCO - [link]CrowdHuman - [link]
Cascade - [link]MAP: 89.3%MAP: 57.3%
Yolo v2 - [link]MAP: 78.6%Close
Faster RCNN - [link]MAP: 73.2%MAP 43.9%MAP 84.95%
“ sizeless is a game changer for transparent and reproducible research. I can now easily run, test and compare my computer vision models to other models and data sets in a few clicks.“
PhD student ETH Zurich