In one sentence
Every CodeGateway account uses the same flat 1.07× service markup, regardless of spend or top-up size. Top-ups carry a 2% service fee that is added to the top-up amount at checkout.
This article answers four questions: what is the markup, how do I compute my price, how is the top-up fee charged, and where can I check usage?
What the flat markup is
CodeGateway is a multi-model API aggregation gateway. The markup is the service multiplier we apply on top of official upstream token prices, covering servers, monitoring, support, and a stability margin.
Every account gets the same multiplier from day one: 1.07×. There are no volume rebates, and no reason to stockpile spend to reach a lower rate.
Example: Claude Sonnet 4.6's official input price is $3 per 1M tokens. At 1.07× you pay $3.21 per 1M input tokens; the $15 per 1M output price becomes $16.05.
How your price is calculated
Your price = official upstream price × 1.07. Examples for common models (USD per 1M tokens):
Model | Official input | Official output | Your price (input / output) |
|---|---|---|---|
Claude Sonnet 4.6 | $3 | $15 | $3.21 / $16.05 |
Claude Haiku 4.5 | $1 | $5 | $1.07 / $5.35 |
Claude Opus 4.6 | $5 | $25 | $5.35 / $26.75 |
When a provider changes official pricing, we adjust the base prices in lockstep; the 1.07× multiplier itself stays the same. The full price list lives on the models and pricing page.
Top-ups and the 2% service fee
Top-ups carry a 2% service fee on the top-up face amount. It is added on top of the amount at checkout; card processing is charged separately by Stripe at the rates shown on the checkout page.
Example: a $10 top-up credits $10 to your balance. The service fee is $0.20, so the checkout total is $10.20 plus card processing. The fee scales with the amount and never changes your per-token price.
Markup and service fee are separate: the markup applies to every API call, while the service fee is charged once per top-up.
Cache billing
Anthropic prompt caching follows Anthropic's official rules: cache writes bill at 1.25× the official input price and cache reads at 0.1× of it. CodeGateway applies the same flat 1.07× markup on top of those official rates.
Using the cache well can cut costs substantially, especially when the system prompt dominates. See the Anthropic Prompt Caching documentation for details.
Where to check usage
Log in to the Dashboard → Overview to see your balance, recent usage, and a per-model cost breakdown. At checkout, the top-up amount and the 2% service fee are shown as separate line items.
FAQ
Q: Do I get a discount for using more?
A: No. Every account pays the same 1.07×, regardless of 90-day spend or top-up size.
Q: Does the markup change over time?
A: No. 1.07× is a public commitment; significant changes are announced 30 days in advance by email or on the site (ToS §12).
Q: Does the 2% top-up fee raise my per-token price?
A: No. The fee is charged once per top-up; every request still bills at 1.07×.
Q: What happens when a provider changes official pricing?
A: We adjust base prices in lockstep. The 1.07× multiplier does not change.
Related docs
Top-up & Billing Guide — top-up flow, refunds, invoices
Error Troubleshooting Guide — 401 / 429 / 500 quick reference
Models & pricing — every model and its live price