Skip to main content
Lium.io logo

Lium.io

Platform Supporter

GPU pods billed per second, from RTX 5090 to B300

Lium.io is the cheapest provider for A100 PCIe 80GB and H100 PCIe today, and under the market median on 21 of its 27 GPUs.

Lium.io is a GPU rental marketplace: independent providers list NVIDIA GPU machines, Lium checks and scores each machine's GPUs, and renters rent them billed per second.

At a glance · prices collected Nov 16, 2026

What's good about Lium.io

  • Per-second billing with no minimum rental and no commitment
  • Public machine-readable pricing for every model and every rentable node
  • Documented CLI, Python SDK and REST API
  • Multiple global locations for deployment
  • Root and Docker-in-Docker access on every pod

Lium.io GPU prices

30 GPU models across 85 configurations. Rank compares Lium.io's cheapest on-demand listing for each GPU with every provider's cheapest.

Lium.io's cheapest on-demand price per GPU against the market median
GPULium.io-60%median+100%Rankvs. medianReport
H100 SXM

8× · secure · Helsinki, Finland

$2.90/hr#19 of 57 providers-9%
H200 SXM

1× · secure · Singapore, Singapore

$3.80/hr#11 of 41 providers-15%
RTX PRO 6000

8× · secure · Seward, United States

$1.49/hr#7 of 41 providers-25%
L40S

8× · secure · Dallas, United States

$0.59/hr#4 of 39 providers-46%
B200

1× · secure · Kansas City, United States

$6.40/hr#12 of 33 providers-15%
HGX B300

8× · secure · Dublin, Ireland

$9.00/hr#18 of 33 providers+0%
RTX A6000

8× · secure · Amsterdam, The Netherlands

$0.40/hr#3 of 26 providers-26%
RTX 5090

8× · secure · Hong Kong, Hong Kong

$0.49/hr#5 of 25 providers-18%
RTX 4090

3× · secure · Atlanta, United States

$0.30/hr#2 of 23 providers-29%
RTX 6000 Ada

1× · secure · Plovdiv, Bulgaria

$0.79/hr#8 of 20 providers-7%
A100 PCIe 80GB

1× · secure · San Antonio, United States

$0.81/hrLowest of 17 providers-45%
H100 PCIe

1× · secure · Almaty, Kazakhstan

$1.90/hrLowest of 15 providers-32%
RTX 3090

6× · secure · Nizhniy Novgorod, Russia

$0.18/hr#7 of 14 providersat median
Tesla T4

1× · secure · Sao Paulo, Brazil

$0.22/hr#4 of 12 providers-44%
RTX 4080

1× · secure · Port Saint Lucie, United States

$0.20/hr#3 of 11 providers-26%
H200 NVL

1× · secure · Beltsville, United States

$3.65/hr#5 of 9 providersat median
RTX 5060 Ti

1× · secure · Riga, Latvia

$0.12/hr#3 of 9 providers-11%
RTX 5080

1× · secure · Irving, United States

$0.19/hr#2 of 9 providers-10%
RTX 4080 SUPER

1× · secure · Port Saint Lucie, United States

$0.22/hr#2 of 8 providers-25%
RTX 5070 Ti

1× · secure · Marseille, France

$0.22/hr#4 of 8 providersat median
RTX PRO 5000

1× · secure · Hanover Park, United States

$0.80/hr#2 of 8 providers-7%
RTX 4060 Ti

2× · secure · Port Saint Lucie, United States

$0.14/hr#5 of 7 providers+12%
RTX 5070

1× · secure · Hornell, United States

$0.17/hr#4 of 6 providers+6%
RTX 3090 Ti

1× · secure · Saint Paul, United States

$0.25/hr#2 of 5 providers-5%
RTX 4070 Ti SUPER

1× · secure · Abu Dhabi, United Arab Emirates

$0.20/hr#2 of 4 providers-10%
RTX A4500

1× · secure · Saint Paul, United States

$0.17/hr#2 of 4 providers-6%
RTX PRO 4000

4× · secure · Pittsburg, United States

$0.27/hr#2 of 4 providers-5%

All configurations

Prices last updated November 16, 2026

Pricing
A100 PCIe 80GB
$0.810/hrsecureSan Antonio, United States
1×
B200
$6.40/hrsecureKansas City, United States
1×
B200
$8.89/hrsecureCharlotte, United States
8×
H100 PCIe
$1.90/hrsecureAlmaty, Kazakhstan
1×
H100 SXM
$2.90/hrsecureHelsinki, Finland
8×
H200 NVL
$3.65/hrsecureBeltsville, United States
1×
H200 NVL
$3.85/hrsecureMontreal, Canada
2×
H200 NVL
$4.75/hrsecurePalm Beach Gardens, United States
4×
H200 SXM
$3.80/hrsecureSingapore, Singapore
1×
H200 SXM
$4.19/hrsecureGwangmyeong, South Korea
8×

What a run costs on Lium.io

24 hours at today's cheapest listing for each GPU, at the size it is listed.

  • 8× H100 SXM for 24 hours

    secure, Helsinki, Finland at $2.90/GPU-hr

    $556.80

    $57.60 under the market median rate

  • 1× H200 SXM for 24 hours

    secure, Singapore, Singapore at $3.80/GPU-hr

    $91.20

    $16.56 under the market median rate

  • 8× RTX PRO 6000 for 24 hours

    secure, Seward, United States at $1.49/GPU-hr

    $286.08

    $96.00 under the market median rate

The market median is every provider's cheapest on-demand price for that GPU.

Billing & terms

Pay-as-you-go
Billed per second at the listed hourly rate from a prepaid balance, with no minimum rental and no commitment.
Secure Tier
The default tier; providers are penalized if a rental is interrupted.
Spot Tier
Cheaper nodes that may be interrupted without compensation.
Regions
Available in Australia, United States, various European countries
Support
Email support@lium.io, or open a private ticket in #tickets on the official Lium Discord server (discord.gg/lium). Documentation at docs.lium.io, with a machine-readable index at docs.lium.io/llms.txt.

Watch out for

  • Marketplace capacity: available GPUs and prices vary by provider and change continuously
  • Spot nodes may be interrupted without compensation
  • Prepaid balance required; pods stop when the balance runs out

Features

Diverse GPU Options
Offers a variety of NVIDIA GPUs including B300, B200, H200, H100, RTX PRO 6000, RTX 5090 and RTX 4090.
Secure and Spot Tiers
Every node is listed as Secure or Spot; Spot nodes are cheaper but may be interrupted without compensation.
Global Availability
Data centers in multiple countries including the US, Europe, and Australia.
Verified Machines
Lium checks every machine's GPUs and scores their performance before it is rented.
GPU Splitting
Nodes can be split so several renters each rent a whole number of a node's GPUs.
CLI, SDK and REST API
Headless CLI and Python SDK with JSON output, a REST API with scoped API keys and per-key USD budgets, and rent-by-spec under a price ceiling.
AI Agent Support
An AI agent with only a shell can start a pod with a time-to-live that removes it automatically, within a set budget; a docs MCP server exposes live GPU prices and availability.
Pod Storage
Templates, persistent volumes, encrypted local volumes, and scheduled backups to S3 with restores.

Company

Published by Lium.io to identify the company behind this listing.

Getting started

  1. Step 1

    Sign Up

    Create an account on lium.io and add an SSH key.

  2. Step 2

    Top Up Balance

    Add funds to your prepaid balance.

  3. Step 3

    Choose a Node

    Pick a GPU node by model, count, location and tier (Secure or Spot).

  4. Step 4

    Deploy a Pod

    Select a template and deploy, from the dashboard, the CLI, or the API.

  5. Step 5

    Connect and Stop

    Connect over SSH; billing stops when the pod is deleted.

Where this page's facts come from

GPU prices
  • Lium Public Nodes Feed (lium.io) · last collected Nov 16, 2026
Profile
Reviewed Oct 27, 2026
Method
How we collect and check prices

Frequently Asked Questions

What GPU types does Lium.io offer?

Lium.io offers various GPU types including Tesla T4, RTX PRO 4000, RTX A6000, RTX A6000, RTX A6000, RTX A6000, RTX A6000, H100 SXM, H100 SXM, H100 SXM, H100 SXM, H100 SXM, H200 SXM, H200 SXM, A100 PCIe 80GB, RTX 5080, RTX 5080, RTX 4080 SUPER, RTX 4070 Ti SUPER, RTX 4070 Ti SUPER, RTX 4080, RTX 4090, RTX 4090, RTX 4090, RTX 4090, RTX 4090, RTX 4090, RTX 4090, RTX 4090, RTX 3070, RTX 3070 Ti, RTX 6000 Ada, RTX 6000 Ada, RTX 6000 Ada, RTX 6000 Ada, HGX B300, HGX B300, HGX B300, HGX B300, RTX 4060 Ti, RTX 4060 Ti, RTX PRO 5000, RTX PRO 5000, RTX A4000, H200 NVL, H200 NVL, H200 NVL, RTX 5060 Ti, RTX 5060 Ti, RTX 5060 Ti, RTX 5060 Ti, RTX 5060 Ti, RTX 5060 Ti, RTX 3090 Ti, RTX 5070 Ti, RTX 5070 Ti, RTX 5070 Ti, RTX A4500, L40S, RTX 5070, RTX 5070, H100 PCIe, RTX 5090, RTX 5090, RTX 5090, RTX 5090, RTX 5090, RTX 5090, RTX 5090, RTX 5090, B200, B200, B200, RTX PRO 6000, RTX PRO 6000, RTX PRO 6000, RTX PRO 6000, RTX PRO 6000, RTX 3090, RTX 3090, RTX 3090, RTX 3090, RTX 3090, RTX 3090, RTX 3090. Check the pricing table above for current availability and pricing.

How do I get started with Lium.io?

Sign Up, Top Up Balance, Choose a Node, Deploy a Pod, Connect and Stop

What are Lium.io's main advantages?

Lium.io's main advantages include: Per-second billing with no minimum rental and no commitment, Public machine-readable pricing for every model and every rentable node, Documented CLI, Python SDK and REST API, Multiple global locations for deployment, Root and Docker-in-Docker access on every pod.

What are Lium.io's limitations?

Lium.io's main limitations include: Marketplace capacity: available GPUs and prices vary by provider and change continuously, Spot nodes may be interrupted without compensation, Prepaid balance required; pods stop when the balance runs out.

Compare Providers

Find the best prices for the same GPUs and models from other providers · All Lium.io alternatives, ranked →

Vast.ai logo

Vast.ai

28 shared GPUs with Lium.io

Compare Prices
VoltageGPU logo

VoltageGPU

24 shared GPUs with Lium.io

Compare Prices
Runpod logo

Runpod

23 shared GPUs with Lium.io

Compare Prices

All Lium.io Comparisons

Lium.io vs Vast.aiLium.io vs VoltageGPULium.io vs RunpodLium.io vs Fluence CloudLium.io vs AquanodeLium.io vs Omega GradientLium.io vs gpu.aiLium.io vs RuncrateLium.io vs Salad CloudLium.io vs Hexgrid CloudLium.io vs ShadeformLium.io vs HyperstackLium.io vs Massed ComputeLium.io vs SesterceLium.io vs SpheronLium.io vs Akash NetworkLium.io vs IO.NETLium.io vs OpenRelayLium.io vs VerdaLium.io vs AceCloudLium.io vs CloudGPULium.io vs DaytonaLium.io vs ModalLium.io vs ParalonCloudLium.io vs QuantaCloudLium.io vs SeewebLium.io vs Amazon AWSLium.io vs CoreWeaveLium.io vs NebiusLium.io vs OblivusLium.io vs Oracle CloudLium.io vs QuickPodLium.io vs SwissGPULium.io vs BasetenLium.io vs CivoLium.io vs CrusoeLium.io vs DigitalOceanLium.io vs fal.aiLium.io vs Google CloudLium.io vs Hugging FaceLium.io vs Impossible CloudLium.io vs LyceumLium.io vs Together AILium.io vs UpCloudLium.io vs Deep InfraLium.io vs FireVMLium.io vs GcoreLium.io vs GPU OutletLium.io vs KoyebLium.io vs Lambda LabsLium.io vs Microsoft AzureLium.io vs Novita AILium.io vs Packet AILium.io vs ScalewayLium.io vs TheAI CloudLium.io vs VultrLium.io vs 1LegionLium.io vs Compute CheapLium.io vs GMI CloudLium.io vs HyperbolicLium.io vs Jarvis LabsLium.io vs Latitude.shLium.io vs OVHcloudLium.io vs Pacific MachinesLium.io vs PaperspaceLium.io vs AircloudLium.io vs Atlas CloudLium.io vs Avian ComputeLium.io vs centronLium.io vs Cudo ComputeLium.io vs Denvr DataworksLium.io vs MeshiveLium.io vs Mulligan ComputeLium.io vs Theta EdgeCloudLium.io vs Thunder ComputeLium.io vs Akamai CloudLium.io vs BeamLium.io vs BentausLium.io vs Beyond.plLium.io vs EcoHashLium.io vs NEOXIS Corp.Lium.io vs Orion AI FactoryLium.io vs Voltage ParkLium.io vs ZettabyteLium.io vs Hot AisleLium.io vs LinuxGPULium.io vs NoctHost

The weekly price report

One email every Monday with the week's notable GPU and LLM API price moves, plus occasional product news. No spam.