Skip to main content

​
Overview

Splox uses pay-as-you-go pricing with no upfront costs, subscriptions, or hidden fees. You only pay for what you use, when you use it.

How Pricing Works

Two simple pricing models:
  1. AI Models & Services - Pay per usage (tokens, generations, seconds)
  2. BYOK (Bring Your Own Key) - Fixed fee per API request ($0.005)
All pricing is transparent with real-time tracking in your dashboard.

​
Pricing Models

​
Pay-Per-Usage Pricing

Use Splox API keys and pay only for what you consume:Language Models (LLMs)Sandboxes
  • Priced per second of compute time
  • vCPU: $0.0000364/sec, Memory: $0.000013/GiB/sec
  • Examples: $0.31 to $1.42 per hour
  • View sandbox pricing →

​
Compare Options

Splox-Managed Keys

Best for:
  • Getting started quickly
  • Testing multiple models
  • Teams without existing API accounts
Pros:
  • Instant access to 600+ models
  • One API key for everything
  • No account setup with providers
  • Unified billing
Pricing:
  • Pay per token/generation/second
  • No fixed fees

BYOK (Your Keys)

Best for:
  • Existing API credits/discounts
  • High-volume usage
  • Enterprise compliance
Pros:
  • Use existing provider credits
  • Direct provider billing
  • Full API key control
  • Transparent costs
Pricing:
  • $0.005 per request
  • Provider costs separate

​
Billing & Usage

​
How Billing Works

1

Pay-As-You-Go

No upfront costs or subscriptions. You only pay for what you use.
2

Real-Time Tracking

Monitor your usage and costs in real-time through the dashboard.
3

Usage Limits

Set spending limits to control costs and avoid unexpected charges.
4

Detailed Invoicing

Get itemized invoices with breakdowns by model, workflow, and date.

​
Token Counting

Tokens are counted using the same tokenization as the model provider:
  • GPT models: ~4 characters = 1 token
  • Claude models: ~4 characters = 1 token
  • Gemini models: ~4 characters = 1 token
Use our token calculator in the dashboard to estimate costs before running workflows.
  • Input tokens: Your prompt + system messages + context + previous conversation
  • Output tokens: The model’s generated response
Output tokens are typically more expensive than input tokens.
Some models (Claude, GPT) support prompt caching:
  • Cache Write: First time sending repeated content (more expensive)
  • Cache Read: Subsequent uses of cached content (much cheaper)
  • Cache TTL: 5 minutes or 1 hour depending on model
Caching can reduce costs by 90% for repeated prompts.

​
Provider Pricing Pages

View detailed pricing for each provider:

OpenAI

GPT-5, GPT-4o, o1, o3, o4-mini

Anthropic

Claude Opus, Sonnet, Haiku

Google Gemini

Gemini 2.5, 2.0, 1.5 series

OpenRouter

200+ models, unified API

Perplexity

Sonar Pro, Sonar Reasoning

Sandbox

Code execution environments

​
FAQ

We accept:
  • Credit cards (Visa, Mastercard, Amex)
  • Debit cards
  • ACH transfers (for enterprise)
All payments are processed securely through Stripe.
Yes! Enterprise customers with high usage (>$1,000/month) can contact us for custom pricing:
  • Volume discounts
  • Dedicated support
  • Custom SLAs
  • Priority access to new models
Contact [email protected] for enterprise pricing.
It depends on your usage:Splox-Managed Keys are cheaper when:
  • You’re testing/experimenting
  • Low-medium volume usage
  • You want simplicity
BYOK is cheaper when:
  • You have existing API credits
  • Very high volume (>100K requests/month)
  • You have provider volume discounts
Use our cost calculator to compare!

​
Need Help?

Contact Sales

Have questions about pricing or need a custom plan?Email: [email protected]Enterprise inquiries: We offer custom contracts, volume discounts, and dedicated support for teams.