Understand HuiLink’s pricing model.
HuiLink uses a transparent, pay-as-you-go pricing model. You are charged per token for both input and output, with discounts for cache hits.
Each model has a per-token price for input and output. Prices are listed per 1M tokens on the Models page.
All prices are in USD. Tokens are counted using the model’s native tokenizer. Both input and output tokens are billed.
When a request matches a cached prompt, you are charged the cache rate instead of the full input rate. This can significantly reduce costs for repeated system prompts or frequent contexts.
For advanced use cases, administrators can configure custom pricing rules using mathematical expressions based on token usage. This enables complex billing scenarios such as volume discounts, tiered pricing, or model-specific multipliers.
input_tokens * 0.000002 + output_tokens * 0.000008Expression-based pricing is configured per channel by administrators. Contact your admin for details.
| Model | Input ($ / 1M tokens) | Output ($ / 1M tokens) | Cached Input ($ / 1M tokens) | Savings with cache |
|---|---|---|---|---|
Usage is deducted from your balance in real-time as requests are processed. You can view your balance and usage history in the dashboard.
No. HuiLink is pay-as-you-go with no monthly commitment. You only pay for the tokens you use.
Tokens are counted by the upstream provider using their native tokenizer. Different models may count tokens differently for the same text.
You can set a spending limit, an expiry date, a model allowlist, an IP allowlist, RPM/TPM rate limits, and a concurrency cap for each key.
Create an account and start using HuiLink today.