Skip to main content

​
How Pricing Works

One account, one API key, one invoice — across every provider. You pay per token, with no monthly minimum and no commitment.

Free Tier

Get started free with 100K tokens per month included. No credit card required.

​
Pay-As-You-Go Pricing

Prices below are per 1M tokens, listed as input / output.
This page shows a representative sample. The Supported Models page carries the full, authoritative price list for all 400+ models — always check there before estimating.
Prices are per 1M tokens. Input and output are billed separately, at different rates.

​
Free Tier

Every SaveGate account includes:
  • 100K tokens/month free
  • Access to all models
  • No credit card required
  • Full API access
  • Real-time analytics
Perfect for:
  • Testing and development
  • Small projects
  • Learning and experimentation

​
Usage-Based Billing

  1. Track Usage: We track your token usage across all models
  2. Monthly Billing: Charges accumulate throughout the month
  3. Auto-Pay: Automatically charged at month-end
  4. Detailed Invoices: View breakdown by model and day
Tokens are pieces of words used for processing. On average:
  • English: ~4 characters per token
  • Code: ~3 characters per token
  • Other languages: Varies by language
Example: “Hello, world!” = ~4 tokensUse our token calculator in the dashboard to estimate costs.
Set up alerts to monitor spending:
  • Daily spending alerts
  • Monthly budget limits
  • Per-model cost tracking
  • Automatic email notifications
We accept:
  • Credit/Debit cards (Visa, Mastercard, Amex)
  • Wire transfer (for enterprise)
  • Crypto payments (coming soon)

​
Enterprise Plans

For large-scale deployments and organizations:

Team Plan

Starting at $500/month
  • Volume pricing on request
  • Team management
  • Priority support
  • Dedicated account manager
  • Custom rate limits

Enterprise Plan

Custom Pricing
  • Negotiated rates
  • SLA guarantees
  • On-premises deployment
  • Custom integrations
  • 24/7 phone support
  • Legal & compliance review

Contact Sales

Get a custom quote for your organization

​
Cost Calculator

Estimate your monthly costs:
Assumptions
  • 10,000 conversations/month, 10 messages each
  • 100 input + 150 output tokens per message
  • Model: gpt-4o-mini — 0.15/0.15 / 0.60 per 1M
Calculation
  • Messages: 10,000 × 10 = 100,000
  • Input: 100,000 × 100 = 10M tokens → 10 × 0.15=∗∗0.15 = **1.50**
  • Output: 100,000 × 150 = 15M tokens → 15 × 0.60=∗∗0.60 = **9.00**
  • Total: $10.50/month
Assumptions
  • 1,000 articles/month
  • 500 input + 2,000 output tokens each
  • Model: claude-sonnet-4-6 — 3/3 / 15 per 1M
Calculation
  • Input: 1,000 × 500 = 0.5M tokens → 0.5 × 3=∗∗3 = **1.50**
  • Output: 1,000 × 2,000 = 2M tokens → 2 × 15=∗∗15 = **30.00**
  • Total: $31.50/month
Assumptions
  • 500 sessions/month, 20 requests each
  • 200 input + 300 output tokens per request
  • Model: gpt-5.4-mini — 0.75/0.75 / 4.50 per 1M
Calculation
  • Requests: 500 × 20 = 10,000
  • Input: 10,000 × 200 = 2M tokens → 2 × 0.75=∗∗0.75 = **1.50**
  • Output: 10,000 × 300 = 3M tokens → 3 × 4.50=∗∗4.50 = **13.50**
  • Total: $15.00/month
Assumptions
  • 2,000 images/month
  • Model: gemini-2.5-flash-image — $0.039 per generated image
Calculation
  • Output: 2,000 × 0.039=∗∗0.039 = **78.00**
  • Total: ~$78/month
Generated images are billed at a flat per-image rate. Text tokens in the same request are billed separately at the model’s normal input/output rates.

​
Billing FAQs

  • Charges accumulate throughout the month
  • Billed on the 1st of each month
  • Free tier resets monthly
  • Invoice sent via email
  • Refunds available for billing errors
  • Contact support within 30 days
  • Unused credits can be refunded
  • Enterprise plans have custom terms
  • Set up budget alerts in dashboard
  • API access continues by default
  • Can set hard limits to stop API calls
  • Email notifications before limits
Yes — volume pricing is agreed per account rather than published as a fixed table, so it can be matched to your actual traffic mix.Email [email protected] with your monthly volume and the models you use, and we will quote you directly.

​
View Your Usage

Track costs in real-time:
  1. Go to Dashboard
  2. Navigate to “Usage & Billing”
  3. View breakdown by:
    • Model
    • Day/Week/Month
    • Application (if using multiple keys)
    • Cost trends

​
Keeping Costs Down

Right-size the model

Start with gpt-4o-mini or gemini-2.5-flash for simple tasks and move up only when quality demands it. The gap between tiers is often 10x or more.

Turn down Gemini thinking

Gemini models reason before answering and bill those tokens as output. Set reasoning_effort on short tasks — see Supported Models for the correct value per model.

Optimize Prompts

Shorter prompts = lower costs. Be concise and specific.

Cache Responses

Cache common responses to avoid duplicate API calls.

Batch Requests

Process multiple items in single requests when possible.