Home

32GB RAM · 16 vCPU$0.18/hr128GB RAM · 32 vCPU$0.62/hr256GB RAM · 48 vCPU$1.12/hr1× L4$0.74/hr1× RTX 4090$1.05/hr2× RTX 4090$2.10/hr1× A100$2.45/hr4× A100$9.20/hr1× H100$3.90/hr8× H100$28.50/hr1× A10$0.88/hr64GB RAM · 64 vCPU$0.34/hr32GB RAM · 16 vCPU$0.18/hr128GB RAM · 32 vCPU$0.62/hr256GB RAM · 48 vCPU$1.12/hr1× L4$0.74/hr1× RTX 4090$1.05/hr2× RTX 4090$2.10/hr1× A100$2.45/hr4× A100$9.20/hr1× H100$3.90/hr8× H100$28.50/hr1× A10$0.88/hr64GB RAM · 64 vCPU$0.34/hr

The market for AI compute

Rent any machine.
Run any model.

GPUs, CPUs and high-RAM machines by the hour, 139+ open and frontier models behind one key — every workload matched to hardware that actually fits, settled on-chain in ETH.

Compute marketETH

1× H100 80GB

$3.90/hr

▲ live pricing
1× L4US-West$0.66/hr
1× RTX 4080EU-Stockholm$0.79/hr
1× A10AP-Tokyo$0.88/hr
View all 16 machines
16
machines
12
open models
10
providers
11
GPU machines

01

What you can do

Everything you need to run open-source AI on rented infrastructure.

02

Call the market. Long or short.

The RAM Index tracks H100 spot — the cheapest live 1×H100 $/hr on the marketplace. Every 24h epoch, pick a side.

Pick a side

LONG if compute closes pricier, SHORT if it closes cheaper — settled against recorded index ticks.

Stake ETH or $RAMDAQ

Your stake is a real on-chain transfer, verified before the position is recorded.

Winners take the pot

User vs user — the losing pool is split among winners pro-rata and payouts hit your wallet automatically.

LONG

Compute closes higher.

▲ bulls split the pot

SHORT

Compute closes lower.

▼ bears split the pot

The 2% fee on every settled pot flows to the treasury — where $RAMDAQ burns automatically. One-sided or flat epochs refund everyone in full.

03

How it works

Rent compute or grab an API key, then use it — in three steps.

1 · Rent a machine or create a key

On the Marketplace, rent a GPU/CPU/RAM machine with ETH — or create a universal API key and top it up in ETH.

2 · Open the Workspace

Your rented machines and your API key both live in the Workspace, ready to use in one place.

3 · Run it

Open a Jupyter notebook (or SSH in) to use a machine, or send prompts to any model with your API key. Stop anytime.

You send a request

A prompt, job, benchmark, or deploy.

Router picks compute

Matches RAM / GPU / CPU to the workload + your strategy.

Runs on a rented machine

A marketplace machine with enough resources.

Open model executes

Llama, Qwen, Mistral, SDXL, Whisper…

Result + cost back

Output, latency, RAM/GPU used, price.

Every action — a prompt, deploy, benchmark or job — is matched to a machine with enough RAM / GPU / CPU to run it.

$RAMDAQ token

Pay in $RAMDAQ for 50% off — and every token burns.

At checkout you choose ETH or $RAMDAQ. Pay in $RAMDAQ and it's 50% cheaper. Every $RAMDAQ spent goes to the RAMDAQ treasury and is burned automatically every 10 minutes — so supply only shrinks as the platform is used.

50% cheaper

Pay for compute, API credits and more in $RAMDAQ at half the ETH price.

Autonomous burn

Treasury $RAMDAQ is burned on-chain every 10 minutes. Deflationary by design — supply only goes down.

Transparent treasury

0xb495…2A973d

Payments land in one public wallet, then burn — all verifiable on-chain.

$RAMDAQ launches soon on Ethereum — pay in ETH today, $RAMDAQ the moment it's live.

04

Two layers, one platform

A compute layer you rent, and an open-source AI layer that runs on it.

Compute layer

Rent the hardware

Raw infrastructure, billed by the hour.

  • Rent CPU, GPU and RAM across providers
  • Launch ready-to-use cloud computers
  • Run heavy jobs straight from the browser
Router
Open-source AI layer

Run the models

Open models, served and compared.

  • Deploy models as private API endpoints
  • Benchmark quality, latency and cost
  • Route requests through one smart API

When you deploy a model, run a benchmark, use the API router, or submit a job, RAMDAQ matches it to a machine with enough RAM / GPU / CPU — automatically.

05

Workload to hardware

How common workloads map to the right kind of machine.

Small chat model

Cheap CPU / RAM machine

Huge Llama / Qwen model

High-RAM or GPU machine

Stable Diffusion

GPU machine (≥16GB VRAM)

Whisper transcription

CPU / GPU machine

Long document prompt

High-RAM machine

Private business workload

Dedicated, high-trust machine

06

FAQ

Straight answers on renting compute, using it, and paying.

I rented a machine — how do I actually use it?
It shows up in the Workspace about a minute after you pay. Click Use it here to open a Jupyter notebook right in the page and run code on the machine's GPU, or copy its SSH command to connect from your own terminal.
Can I use the machine from my own computer?
Yes. Each machine shows an SSH command in the Workspace — copy it into your terminal (Windows Terminal, macOS Terminal, Linux) to log in and run anything, upload files, and more.
What's the difference between renting a machine and the API keys?
Renting a machine gives you a whole computer (GPU/CPU/RAM) you control via Jupyter or SSH. An API key lets you call ready-made models (Claude, Llama, Qwen…) over an OpenAI-compatible endpoint and pay per token — no machine to manage.
How do I use my API key?
Create a key under Marketplace → Model APIs (or in the Workspace), top it up in ETH, then either use the built-in prompt box or call the endpoint from your code — Authorization: Bearer <key>, OpenAI-compatible.
How does Long / Short the RAM Index work?
Each 24h epoch records the RAM Index (H100 spot $/hr) at open and close. Stake ETH or $RAMDAQ on LONG or SHORT; if the index closes above the open, longs split the short pool pro-rata (minus a 2% fee that burns $RAMDAQ) — and vice versa. It's user vs user: one-sided or flat epochs refund everyone, and payouts land in your wallet automatically at settlement.
How do I pay, and when am I billed?
Everything is paid in ETH with your Ethereum wallet (Robinhood Wallet, MetaMask, or any browser wallet). Machines bill per hour while active — click Stop to end charges. API keys are prepaid and charged per token as you call models; each top-up is verified on-chain.
Are the machines real?
Yes — the Marketplace lists live RunPod GPUs with real prices, and renting launches a real pod with Jupyter + SSH. If RunPod has nothing free, it falls back to a clearly-labelled demo catalog.
Can I pay with the $RAMDAQ token?
Yes — at checkout you choose ETH or $RAMDAQ, and paying in $RAMDAQ is 50% cheaper. (Launching soon — ETH works today.)
What happens to $RAMDAQ when I spend it?
It goes to the RAMDAQ treasury and is burned autonomously every 10 minutes on-chain — so $RAMDAQ supply only shrinks as the platform is used.

The market is open

Run open-source AI on rented compute.

Pick a machine, deploy a model, benchmark the trade-offs, and serve it through one API — all from your dashboard.