Sign in Request access

NVIDIA GPUs · Secure and Community

Your workload.Your budget.The right compute.

The GPU your model needs. A budget you never pay past.

Request access Try Smart Launch

The Node Lodge console launcher: an H100 80GB in Montreal, Canada, an 8-hour budget and its guaranteed maximum cost.

Sized to your model
A budget on every session
Secure or Community
CLI, SDKs and MCP

Smart Launch

Type a model. See what fits.

smart-launch · try it

Smart Launch · the real engine, in your browser.

$ llama 3.1 70b 4-bit 32k

Llama 3.1 70B Instruct · 4-bit · 32,768 tokens · 1 at a time

Model weights 32.9 GiB

KV cache 10.0 GiB

Overhead margin 8.6 GiB

CUDA runtime 1.0 GiB

Needs 52.5 GiB

B300 270 GB fits

B200 180 GB fits

H200 141GB SXM 141 GB fits

RTX PRO 6000 96 GB fits

H100 80GB SXM 80 GB fits

H100 80GB PCIe 80 GB fits

A100 80GB SXM 80 GB fits

L40S 48 GB fits on 2

L40 48 GB fits on 2

RTX A6000 48 GB fits on 2

RTX A4000 16 GB fits on 4

Medium confidence. Same input, same answer.

Also: Llama 3.3 70B Instruct

Your turn: type a model, or help.

Budget Guard
Set a budget. Never pay past it.
Try it
Developers and agents
CLI, SDKs, MCP. Hard caps on every key.
Try the CLI
Secure or Community
Your data decides the tier.
Pick yours

GPU Cloud

Ampere to Blackwell.

Explore GPU Cloud

BlackwellDatacenter
B300
Frontier training, long context
VRAM
270 GB
From
$10.45/h
- France
- United Kingdom
- United States
BlackwellDatacenter
B200
Training and inference at scale
VRAM
180 GB
From
$9.35/h
- Israel
- United States
HopperDatacenter
H200 141GB SXM
Large models in one GPU
VRAM
141 GB
From
$5.94/h
- Canada
- Finland
- France
- United States
HopperDatacenter
H100 80GB SXM
Training and fast inference
VRAM
80 GB
From
$4.95/h
- Canada
- Finland
- United States
HopperDatacenter
H100 80GB PCIe
Inference and fine-tuning
VRAM
80 GB
From
$2.89/h
- Canada
AmpereDatacenter
A100 80GB SXM
Proven training workhorse
VRAM
80 GB
From
$2.01/h
- United States
BlackwellDatacenter
RTX PRO 6000
Large models, rendering
VRAM
96 GB
From
$2.16/h
- Canada
- Spain
- United Kingdom
- United States
Ada LovelaceDatacenter
L40S
Inference, fine-tuning, graphics
VRAM
48 GB
From
Not in stock
- Finland
Ada LovelaceDatacenter
L40
Inference, rendering, video
VRAM
48 GB
From
$1.11/h
- Canada
AmpereDatacenter
RTX A6000
Fine-tuning and rendering
VRAM
48 GB
From
$0.57/h
- Canada
AmpereDatacenter
RTX A4000
Small models and batch jobs
VRAM
16 GB
From
$0.17/h
- Norway

Lowest price per GPU-hour on Nov 9, 2026, in US dollars. Billed per minute.

Use cases

What teams run.

LLM inference · a chat completion, streamed word by word

Request access.

Tell us what you will run. Your access arrives by email.

Request access Sign in