GPU instances for generative AI

Run text, image, video, audio, and multimodal generative AI workloads on GPU containers, VMs, or bare metal. Choose your hardware, provider, and region with transparent pricing and zero egress fees.

DEPLOY GPU

REQUEST GPU CLUSTER

GPU

GPU instance type: h200

H200

Fluence

$2.56/hr

CoreWeave

Core Weave

$6.30/hr

AWS

$7.90/hr

Google Cloud

$10.84/hr

$10.84$

Available GPUs for generative AI workloads

Compare available containers, VMs, and bare-metal configurations for language, visual, audio, and multimodal applications.

Certified for compliance with GDPR, ISO 27001,

SOC2 standards, top-tier facilities and

high-performance servers

All

Container

Bare Metal

H100

80 GB

Info

80 GB RAM

Deployment

Container

Best fit

Production LLM generation

Price

$1.24/hr

DEPLOY GPU

H200

141 GB

Info

141 GB RAM

Deployment

Container

Best fit

Large-context LLMs

Price

$2.56/hr

DEPLOY GPU

L40S

48 GB

Info

48 GB RAM

Deployment

Container

Best fit

Multimodal generation

Price

$1.27/hr

DEPLOY GPU

A100

80 GB

Info

80 GB RAM

Deployment

Container

Best fit

Custom GenAI workloads

Price

$0.96/hr

DEPLOY GPU

RTX 4090

24 GB

Info

24 GB RAM

Deployment

Container

Best fit

Image generation

Price

$0.53/hr

DEPLOY GPU

H100

80 GB

Info

80 GB RAM

Deployment

Best fit

Persistent LLM applications

Price

$2.29/hr

DEPLOY GPU

H200

141 GB

Info

141 GB RAM

Deployment

Best fit

Memory-heavy GenAI

Price

$3.10/hr

DEPLOY GPU

A100

80 GB

Info

80 GB RAM

Deployment

Best fit

Custom model stacks

Price

$1.83/hr

DEPLOY GPU

8× H100

640 GB total

Info

640 GB total

Deployment

Bare Metal

Best fit

Multi-GPU GenAI

Price

$18.42/hr

Request capacity

All

Container

Bare Metal

H100

80 GB

Info

80 GB RAM

Deployment

Container

Best fit

Production LLM generation

Price

$1.24/hr

DEPLOY GPU

H200

141 GB

Info

141 GB RAM

Deployment

Container

Best fit

Large-context LLMs

Price

$2.56/hr

DEPLOY GPU

L40S

48 GB

Info

48 GB RAM

Deployment

Container

Best fit

Multimodal generation

Price

$1.27/hr

DEPLOY GPU

A100

80 GB

Info

80 GB RAM

Deployment

Container

Best fit

Custom GenAI workloads

Price

$0.96/hr

DEPLOY GPU

RTX 4090

24 GB

Info

24 GB RAM

Deployment

Container

Best fit

Image generation

Price

$0.53/hr

DEPLOY GPU

H100

80 GB

Info

80 GB RAM

Deployment

Best fit

Persistent LLM applications

Price

$2.29/hr

DEPLOY GPU

H200

141 GB

Info

141 GB RAM

Deployment

Best fit

Memory-heavy GenAI

Price

$3.10/hr

DEPLOY GPU

A100

80 GB

Info

80 GB RAM

Deployment

Best fit

Custom model stacks

Price

$1.83/hr

DEPLOY GPU

8× H100

640 GB total

Info

640 GB total

Deployment

Bare Metal

Best fit

Multi-GPU GenAI

Price

$18.42/hr

Request capacity

Deploy in the environment your generative AI stack needs

GPU Containers

Run packaged language, image, audio, or video models in a repeatable environment with less infrastructure setup.

• Fast workload launches • Reproducible environments • Packaged model runtimes

Deploy container

Deploy

GPU VMs

Control the operating system, drivers, model files, storage, ports, and supporting application services.

• Custom software stacks • Persistent applications • Private model services

Launch VM

Bare Metal

Use full nodes for sustained GPU utilization, large-memory models, or tightly coupled multi-GPU pipelines.

Dedicated hardware for maximum control, direct hardware access, stronger isolation, and high-utilization inference workloads.

• Dedicated GPU resources • Direct hardware control • Multi-GPU deployments

Request bare metal

Request

From language models to multimodal generation

Language and code generation

Power chat, content creation, summarization, search experiences, coding tools, and internal AI assistants.

Best fit: H100 / H200 / A100

Image and creative production

Generate product visuals, design concepts, marketing assets, image variations, and other visual content.

Best fit: L40S / RTX 4090 / A100

Video, voice, and audio generation

Run video synthesis, speech generation, voice applications, music models, and audio-production pipelines.

Best fit: L40S / H100 / H200

Multimodal products

Build applications that process or generate combinations of text, images, audio, video, and other media.

Best fit: H100 / H200 / L40S

More control over generative AI infrastructure

Clearer compute economics

Compare hourly GPU prices and transfer generated outputs without additional egress fees.

Clearer compute economics

Compare hourly GPU prices and transfer generated outputs without additional egress fees.

Hardware matched to each modality

Choose GPU memory and performance for language, image, audio, video, or multimodal workloads.

Hardware matched to each modality

Choose GPU memory and performance for language, image, audio, video, or multimodal workloads.

Hardware matched to each modality

Choose GPU memory and performance for language, image, audio, video, or multimodal workloads.

Infrastructure on your terms

Run your own model and application stack without depending on a proprietary generative AI platform.

Infrastructure on your terms

Run your own model and application stack without depending on a proprietary generative AI platform.

Capacity across providers

Compare available hardware, regions, configurations, and rates through one GPU marketplace.

Request a custom GPU cluster

Need a different GPU setup or configuration? Send us your requirements and our team will get back to you.

Need a different GPU setup or configuration?

Send us your requirements and our team will get back to you.

High-performance GPUs across trusted global locations

Run generative AI workloads on provider-operated infrastructure across multiple regions. Review the location, provider, hardware configuration, and available compliance information before deployment.

FAQ

What generative AI workloads can I run on Fluence?

Which GPU should I choose for generative AI?

When should I use a container, VM, or bare metal?

Does Fluence provide models or a managed generative AI API?

How is generative AI GPU cost calculated?

What generative AI workloads can I run on Fluence?

Which GPU should I choose for generative AI?

When should I use a container, VM, or bare metal?

Does Fluence provide models or a managed generative AI API?

How is generative AI GPU cost calculated?

Show more

Launch your generative AI stack on Fluence

Bring your models and application environment. Choose containers, VMs, or bare metal with transparent pricing, zero egress fees, and infrastructure freedom.

DEPLOY GPU

REQUEST GPU CLUSTER