Fluidstack
GPU cloud for AI training and inference, with H100, A100 and newer NVIDIA hardware.
Introduction
Fluidstack provides large GPU clusters and on-demand instances for training foundation models and running inference at scale. Hardware includes H100, A100, H200 and GB200, managed with Slurm or Kubernetes. It lists a 99% uptime guarantee and 24/7 support.
- Added on
- Nov 2, 2026
- Website
- fluidstack.io
- Pricing model
- Paid
- Company
- FluidStack
Product information
What is Fluidstack?
Fluidstack is an AI cloud platform that supplies NVIDIA GPU capacity for enterprises training models and serving inference.
How to use Fluidstack?
Reserve a large GPU cluster or launch on-demand instances, choose managed Kubernetes or Slurm, and contact their engineers when you need help.
Core features of Fluidstack
- Thousands of NVIDIA GPUs
- Managed Slurm and Kubernetes
- Large training clusters
- On-demand GPU instances
- 24/7 support
- 99% uptime guarantee
Use cases of Fluidstack
- 1Training large AI models
- 2Running inference at scale
- 3Managing GPU clusters
- 4Speeding up ML product development
Frequently asked questions about Fluidstack
Which GPUs does Fluidstack offer?+
NVIDIA H100, A100, H200 and GB200.
How fast can I launch Fluidstack GPU instances?+
On-demand instances can be launched in under 5 minutes.
What support does Fluidstack provide?+
24/7 support with a 15-minute response time.
Which category is Fluidstack in?+
Fluidstack is listed under AI Code Editor.