Skip to main content

​
Overview

Baseten provides a generous $30 in trial credits to new users. Unlike token-based pricing, Baseten charges by compute time, giving you flexibility to run any supported model from their extensive library.

Trial Credits

$30 in free credits

Pricing Model

Pay by compute time

​
Available Models

Baseten offers access to any model in their extensive library. You pay based on the compute time used rather than per token, which can be more cost-effective for certain workloads.
Browse the complete model library at baseten.co/library

​
Model Categories

  • Language Models: Various sizes and architectures
  • Vision Models: Image generation and analysis
  • Audio Models: Speech and audio processing
  • Custom Models: Deploy your own models

​
Getting Started

​
1. Create an Account

Visit app.baseten.co and sign up for a free account to receive your $30 in credits.

​
2. Deploy a Model

Choose a model from the library or deploy your own.

​
3. Make API Calls

import requests

url = "https://model-<model-id>.api.baseten.co/production/predict"
headers = {
    "Authorization": f"Api-Key {YOUR_BASETEN_API_KEY}"
}

data = {
    "prompt": "What is the capital of France?",
    "max_tokens": 100
}

response = requests.post(url, headers=headers, json=data)
print(response.json())

const response = await fetch(
  'https://model-<model-id>.api.baseten.co/production/predict',
  {
    method: 'POST',
    headers: {
      'Authorization': `Api-Key ${process.env.BASETEN_API_KEY}`,
      'Content-Type': 'application/json'
    },
    body: JSON.stringify({
      prompt: 'What is the capital of France?',
      max_tokens: 100
    })
  }
);

const result = await response.json();
console.log(result);

curl -X POST https://model-<model-id>.api.baseten.co/production/predict \
  -H "Authorization: Api-Key YOUR_BASETEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "What is the capital of France?",
    "max_tokens": 100
  }'

​
Pricing Model

Baseten charges by compute time rather than tokens. This means you’re billed for the actual GPU/CPU time your model uses, which can be more economical for certain use cases.

​
Advantages of Compute-Time Pricing

  • Predictable costs for batch processing
  • Cost-effective for models with variable token generation
  • Flexible for custom model deployments

​
Use Cases

  • Model Deployment: Deploy and test custom models
  • Batch Processing: Run large-scale inference jobs
  • API Integration: Build production applications with reliable endpoints
  • Model Comparison: Test different models to find the best fit

​
Resources

Baseten Platform

Access the platform

Model Library

Browse available models
Monitor your compute time usage to maximize your $30 credit allocation. Different models have different compute costs.

Build docs developers (and LLMs) love

Get started for freeTalk to us