P2P marketplaces Los Angeles, CA · Live data

Vast.ai.

P2P marketplace — list consumer or datacenter GPUs at your own per-hour bid. Huge renter demand, weekly BTC/USD payouts.

At a glance

Business model: P2P marketplaces
Tier: Marketplace
Headquarters: Los Angeles, CA
Mining: Allowed

When to pick Vast.ai

Best for

Cost-sensitive experimentation and hobbyist training.
Batch inference where occasional host swaps are fine.
Anyone willing to filter hosts by reliability score.

Avoid for

Production inference with strict latency SLAs.
Multi-host distributed training — hosts don't share infrastructure.
Workloads with data-residency or compliance requirements.

Price history

Daily median on Vast.ai's top GPUs.

GPUs available on Vast.ai

GPU	Tier	VRAM	$/hr	Range	Offers
Nvidia RTX 4070 Ti SUPER	consumer	16GB	$0.069/hr	$0.069/hr–$0.069/hr	1	Compare →
Nvidia RTX 5070 Ti	consumer	16GB	$0.12/hr	$0.12/hr–$0.12/hr	1	Compare →
Nvidia GeForce RTX 3090	consumer	24GB	$0.12/hr	$0.12/hr–$0.12/hr	1	Compare →
Nvidia GeForce RTX 5080	consumer	16GB	$0.13/hr	$0.13/hr–$0.13/hr	1	Compare →
Nvidia CMP 170HX	consumer	—	$0.20/hr	$0.20/hr–$0.20/hr	1	Compare →
Nvidia RTX PRO 4500 Blackwell	workstation	32GB	$0.35/hr	$0.35/hr–$0.35/hr	1	Compare →
Nvidia GeForce RTX 4090	consumer	24GB	$0.44/hr	$0.44/hr–$0.44/hr	1	Compare →
Nvidia GeForce RTX 5090	consumer	32GB	$0.47/hr	$0.47/hr–$0.47/hr	1	Compare →
Nvidia A100 PCIe	datacenter	40GB	$0.47/hr	$0.47/hr–$0.47/hr	1	Compare →
Nvidia RTX 6000 Ada	workstation	48GB	$0.54/hr	$0.54/hr–$0.54/hr	1	Compare →
Nvidia RTX PRO 5000 Blackwell	workstation	48GB	$0.73/hr	$0.73/hr–$0.73/hr	1	Compare →
Nvidia RTX PRO 6000 S	workstation	96GB	$1.13/hr	$0.99/hr–$1.27/hr	2	Compare →
Nvidia RTX PRO 6000 MaxQ	workstation	96GB	$1.20/hr	$1.20/hr–$1.20/hr	1	Compare →
Nvidia RTX 6000D		—	$1.26/hr	$1.26/hr–$1.26/hr	2	Compare →
Nvidia RTX PRO 6000 WS	workstation	96GB	$1.30/hr	$1.30/hr–$1.30/hr	1	Compare →
Nvidia H200	datacenter	141GB	$2.95/hr	$2.95/hr–$2.95/hr	1	Compare →

Compare with peers

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

TensorDock

80% host

Catalog-only — no live pricing yet

P2P marketplace with 100+ GPU models across home + datacenter hosts. Often the cheapest on-demand rates in the long tail.

Cudo Compute

80% host

Cheapest: $0.88/hr · Nvidia L40S · 2 GPUs

Descended from Cudo Miner — hosts install the Cudo Compute Provider agent and earn on completed compute jobs. USD or BTC payouts.

Clore.ai

85% host

Cheapest: $0.0075/hr · Nvidia GTX 1050 Ti · 82 GPUs

P2P GPU marketplace. Host lists a server, pays out in CLORE, BTC, or USD; renters pay per-hour for on-demand or spot capacity.

Lium

Cheapest: $0.12/hr · Nvidia RTX A4000 · 18 GPUs

GPU marketplace — independent providers list machines and set their own prices, so rates move with competition rather than a fixed rate card. Renters top up ...

Frequently asked

How is Vast.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 Vast.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 Vast.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 Vast.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.