JarvisLabs
India-based GPU cloud with per-minute billing for AI training and inference.
About
JarvisLabs is a GPU cloud provider focused on AI developers, with data centers in India. It rents GPU virtual machines, framework templates, managed model endpoints, and a token-priced model API, billing by the minute so short experiments do not incur hourly minimums.
Instances come with JupyterLab, VS Code Web, and SSH access plus prebuilt PyTorch and TensorFlow templates, and the catalog spans NVIDIA H200, H100, RTX Pro 6000, A100, and L4 cards. The bootstrapped company reports more than 27,000 AI developers and 50 million GPU hours served, and it positions itself as a low-cost option for teams priced out of hyperscaler GPUs.
What it can do
- GPU VMs with per-minute billing and pause/resume
- Managed endpoints and token-priced model APIs
- JupyterLab, VS Code Web, and SSH access
- VPC networking and reserved capacity for teams
Setup
Create an account at jarvislabs.ai, top up credits, pick a GPU and template, and launch.
Prerequisites
A JarvisLabs account with topped-up credits.
Use cases
Works with
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 115; Trending score 66/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.