What Tablif read on Calibra Dataset Intelligence's site
Calibra Dataset Intelligence describes itself as: “Calibra Dataset Intelligence”
- Price model: free tier
- Free tier: Free forever
- Stage: Live, no waitlist
What it is
Calls itself “robotics dataset intelligence layer”.
- Calls itself: robotics dataset intelligence layer
“Calibra is a robotics dataset intelligence layer.”
Where it puts itself
Rules out “upload”.
- Says it will not: upload
“No upload, account, or API key required”
What it costs
A free tier, free forever.
- Free plan: free forever
“Install Calibra locally or inspect a public LeRobot dataset in the browser. Free, no account needed.”
How it runs
A command-line tool. Runs locally and works offline. Distributed through the PyPI. Built with GitHub Pages.
- Kind: cli
- Runs where: on the device, offline
- Distributed via: PyPI
- Built with: GitHub Pages
“$ pip install calibra-robotics”
“Run everything locally, inspect the reasoning, and export the result for your training pipeline.”
What it works with
Names ACT, BC-MLP and 5 more on its pages.
- Names: ACT · BC-MLP · Diffusion Policy · +4
“Curation gains hold up across a broader 5-seed sweep on 3 datasets and 3 policy families (BC-MLP, ACT, Diffusion Policy).”
“Calibra v0.11 makes robotics dataset analysis more accurate and reliable.”
What it offers as proof
Live, no waitlist. Try it without signing up.
- Trial asks for: nothing
“$ pip install calibra-robotics”
“No upload, account, or API key required”
Sources
Tablif read these pages, last on 10 Oct 2026. Everything above is what the product says about itself; Tablif does not verify every claim.
Similar products in Tablif Town
- Ai Tools Finder: Discover the Right AI Tool for Every Task
- FLUX 3 Action: An open-weight AI model for predicting robot actions
- ElideDB. Database for Physical AI: Future of video memory for Physical AI
- SkillsCat: Discover, share, and install AI agent skills
- Kodro: Code robots in an offline Python learning simulator
- Dataset Finder: Your AI Training Data Workspace.
- Deepen Refinery: Real-world training data for Physical AI, verified first
- SimTooReal: Train, validate, and ship robot policies from sim to hardware