Model Pricing
Using Claude Opus 4.7 as an example
What is a token?
A token is the basic unit of text that the model processes. Tokens are not the same as characters or words — they are text fragments determined by the model’s own tokenizer.| Text Example | Approx. Characters | Approx. Tokens |
|---|---|---|
| English word | 4–5 characters | ~1 token |
| Chinese character | 1 character | ~1.5–2 tokens |
| 1,000-word English article | ~5,000 characters | ~750 tokens |
| 1,000-character Chinese article | ~1,000 characters | ~1,500–2,000 tokens |
These are estimates. Different providers such as OpenAI, Google, and Anthropic have slightly different token calculation rules.
Token Types and Billing Rules (Claude Opus 4.7)
Input Tokens (Prompt) — $5.5 / million
Everything you send to the model — system prompt, conversation history, user messages, uploaded documents — all counts as input tokens.Output Tokens (Completion) — $27.5 / million
The text the model generates in response. Output tokens cost 5× more than input tokens, so keeping responses concise has a meaningful impact on cost.Cached Tokens — Billed at 0.25× input rate
Prompt content served from cache is charged at only 25% of the normal price. Useful for long system prompts or repeated context, and can significantly reduce costs.Cost Estimation Examples
| Scenario | Input | Output | Estimated Cost |
|---|---|---|---|
| Short conversation | 1K tokens | 500 tokens | ~$0.020 |
| Medium task | 10K tokens | 3K tokens | ~$0.138 |
| Large codebase analysis | 100K tokens | 10K tokens | ~$0.825 |
Top-Up
After logging into your account on the Routerlink model page, click your avatar in the top-right corner, then click “Top-Up.”
Platform Fee
| Payment Method | Fee | Note |
|---|---|---|
| Credit card | 0 | minimum ≥ 500 Credits |
| Alipay | 0 | minimum ≥ 500 Credits |
| Cryptocurrency (WAI) | A small amount of BNB gas is required. | minimum ≥ 200 Credits |
Each user also receives 200 free credits per day.
Billing
You can view detailed Billing, Logs, and Usage information at here.