GPU cloud marketplace

Powerful GPUs, rented by the second.

Spin up a GPU from a marketplace of idle hardware — from RTX 3090s to A100s — at a fraction of cloud prices. No waitlist, no application. Up and running in under two minutes.

Free credits · no card required
Pay as you go Private & isolated Per-second pricing

RTX 4090 · 24GB

hive-node-3 · Verified host

₹48/hr

● available

98%

uptime

24 GB

VRAM

1 Gbps

network

$ ghive create --gpu rtx4090 --image pytorch

✓ matched hive-node-3 · ₹48/hr

✓ container up · ssh ready in 38s

→ ssh ghive@node-3.ghive.ai

₹12/hr

GPUs from

11+

GPU types priced

5

regions

Per-second

billing

Free credits No card to start Cancel anytime

Works with the tools you already use

PyTorchTensorFlowJAXHugging FaceCUDAJupytervLLMOllamaComfyUIDockerStable DiffusionTritonPyTorchTensorFlowJAXHugging FaceCUDAJupytervLLMOllamaComfyUIDockerStable DiffusionTriton

How it works

From signup to GPU in minutes

1. Add credits

Top up in a few clicks. Get free credits on signup — no card required to start.

2. Pick a GPU

Filter by model, VRAM, price and reliability. Rent on-demand or grab a cheaper interruptible instance.

3. Run your job

SSH or Jupyter in seconds with a pre-baked PyTorch image. Billed per second. Stop anytime.

Live marketplace

Popular GPUs, transparent rates

RTX 3060Verified

12 GB · ap-south

Inference, small fine-tunes, learning

₹12/hr

97% uptime

Rent
RTX 3090Unverified

24 GB · ap-south

13B fine-tunes, Stable Diffusion

₹24/hr

95% uptime

Rent
RTX 4090Verified

24 GB · ap-south

Fast fine-tunes, serving, heavy SD/LLM

₹48/hr

98% uptime

Rent

Use cases

One marketplace, every GPU workload

Fine-tune LLMs

LoRA/QLoRA and full fine-tunes on 24GB–80GB cards without the cloud bill.

Serve inference

Spin up vLLM or Ollama and serve open models with low latency.

Image & video gen

Stable Diffusion, ComfyUI and diffusion pipelines on fast consumer GPUs.

Research & notebooks

Jupyter in seconds for experiments, papers, and coursework.

Batch & training jobs

Cheap interruptible capacity for checkpoint-friendly training runs.

Build AI products

Prototype to production on the same rupee-billed marketplace.

Earn with GHive

Your idle GPU is leaving money on the table

Got a gaming rig, an ex-mining setup, or spare datacenter capacity? Install the GHive Host app, click Start, and earn while you sleep — a transparent rate per GPU type, paid out to you.

  • Taskbar app — one click to start or stop hosting
  • GHive sets a fair fixed rate for your GPU
  • Track hours & earnings in your dashboard
  • Reliable payouts, fully automated
Become a host

Estimated monthly earnings

RTX 3090₹4,000 – ₹9,000
RTX 4090₹8,000 – ₹18,000
A100 40GB₹25,000 – ₹55,000

Illustrative, based on typical utilization. Actual earnings vary with demand and uptime.

Trust & security

Serious infrastructure, sensible safeguards

Cheap doesn't mean careless. The controls that matter, built in.

Isolated workloads

Every rental runs in its own hardened, unprivileged container — never sharing your data with other tenants.

Private by default

Storage is wiped when you stop an instance, and traffic is encrypted end to end.

Prepaid, no surprises

Per-second billing from a prepaid wallet with auto-stop. You never owe more than you load.

Verified hosts

GPUs are benchmarked and spec-checked; reliability scores and trust tiers keep quality high.

Clear invoices

Itemized billing and receipts for every rental — clean for your accounting.

You're in control

Stop any instance instantly. Choose verified or datacenter hosts for sensitive jobs.

FAQ

Questions, answered

GPUs start from about ₹10/hr. GHive publishes one transparent rate per GPU type, billed per second from your prepaid wallet — no setup fees, no egress traps, no minimums.

Better GPUs, less cost. Start building.

Join the waitlist-free marketplace. Get free credits on signup and rent your first GPU today.

No credit card · free credits · cancel anytime