API - SupaScale

Get Started in 3 Simple Steps

From selection to execution in minutes

1

Choose your GPU

Select from our available GPUs based on your memory requirements and budget.

Select GPU & VRAM Size

NVIDIA GeForce RTX 4090

22.5GB Available VRAM

$2.5/hr

10% utilized

2

Choose your AI model

Pick from popular open-source models like Llama, Mistral, or upload your own.

Compatible Models

Select a model

Compatible models will appear based on your GPU selection

3

Connect API in your code

Integrate our simple API into your application and start computing instantly.

Sign up to get your API key

Get Started Free

API Documentation

Simple REST API for powerful GPU computing

Quick Start

1. Create a session

curl -X POST https://api.loomo.io/v1/sessions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "gpu_id": "rtx-4090",
    "model": "llama2-7b"
  }'

2. Send completion request

curl -X POST https://api.loomo.io/v1/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "session_id": "sess_abc123",
    "prompt": "Explain quantum computing",
    "max_tokens": 150
  }'

3. Get response

{
  "id": "comp_xyz789",
  "object": "completion",
  "choices": [{
    "text": "Quantum computing leverages...",
    "finish_reason": "length"
  }],
  "usage": {
    "prompt_tokens": 3,
    "completion_tokens": 150,
    "total_tokens": 153
  }
}

SDKs & Examples

Python SDK

import loomo

client = loomo.Client(api_key="YOUR_API_KEY")

# Create session
session = client.sessions.create(
    gpu_id="rtx-4090",
    model="llama2-7b"
)

# Generate completion
response = client.completions.create(
    session_id=session.id,
    prompt="Explain quantum computing",
    max_tokens=150
)

print(response.choices[0].text)

Node.js SDK

const Loomo = require('loomo');

const client = new Loomo({
  apiKey: 'YOUR_API_KEY'
});

async function generateText() {
  const session = await client.sessions.create({
    gpu_id: 'rtx-4090',
    model: 'llama2-7b'
  });

  const completion = await client.completions.create({
    session_id: session.id,
    prompt: 'Explain quantum computing',
    max_tokens: 150
  });

  console.log(completion.choices[0].text);
}

Installation

pip install loomo

npm install loomo

Simple, Transparent Pricing

Pay only for what you use. No hidden fees.

$0.25 - $4.75

per hour

Based on GPU selection

$0.001

per 1K tokens

Competitive pricing

Free

API calls

No setup or monthly fees

Ready to Scale Your AI?

Join thousands of developers who trust Loomo for their GPU computing needs

Get Started Free View Documentation

SupaScale API