First-party clouds San Francisco, CA

Novita AI.

AI and agent cloud platform providing on-demand GPU rentals for training and inference workloads with high performance and cost efficiency.

At a glance

Business model: First-party clouds
Tier: First-party cloud
Trust tier: Tier 2
Headquarters: San Francisco, CA
Payout: N/A (first-party)

When to pick Novita AI

Best for

Distributed training where InfiniBand interconnect matters.
Regulated workloads requiring SOC 2 / HIPAA.
Teams that need real human support and SLA-backed uptime.

Avoid for

One-off short workloads — the per-hour premium adds up.
Hobbyist single-GPU jobs that don't need premium support.

Price history

Daily median on Novita AI's top GPUs.

GPUs available on Novita AI

GPU	Tier	VRAM	$/hr	Range	Offers
Nvidia RTX 4090	consumer	24GB	$0.18/hr	$0.18/hr–$0.18/hr	1	Compare →
Nvidia L40S	datacenter	48GB	$0.28/hr	$0.28/hr–$0.28/hr	1	Compare →
Nvidia RTX 5090	consumer	32GB	$0.36/hr	$0.36/hr–$0.36/hr	1	Compare →

Compare with peers

Other providers in the same bucket — quick way to sanity-check pricing before committing.

DeepInfra

Tier 1

Cheapest: $2.79/hr · Nvidia B200 · 1 GPUs

First-party AI cloud — DeepInfra also rents B200 GPU instances on demand (1x/2x/4x/8x configs). Same provider as the hosted-inference API, just a different s...

AceCloud

Cheapest: $0.85/hr · Nvidia Tesla V100 · 3 GPUs

Hyperstack

Cheapest: $0.32/hr · Nvidia Nvidia RTX A4000 · 6 GPUs

Lambda

Cheapest: $0.75/hr · Nvidia A10 · 7 GPUs

Frequently asked

How is Novita AI billing measured?

Most providers bill per-second once the instance is running, with a small minimum (often 60 seconds). Some first-party clouds round up to the minute. Headline $/hr is the right comparison unit.

Which regions does Novita AI offer GPUs in?

P2P marketplaces aggregate hosts worldwide so region varies per offer. First-party clouds and hyperscalers expose explicit region pickers (US, EU, APAC). Filter on the provider's site after clicking through if region matters.

Does Novita AI offer an SLA?

Hyperscalers (AWS, GCP, Azure, Oracle) publish formal SLAs. First-party clouds (Lambda, CoreWeave) offer support contracts. P2P marketplaces and decentralized networks have no SLA — uptime depends on the individual host.

What's the refund / cancellation policy on Novita AI?

Per-second billing means you only pay for compute used — stop the instance and billing stops. Pre-paid credits and committed-use discounts have provider-specific terms; check the provider's billing docs before pre-paying.