Pricing.
Pay only the per-million-token rate × actual usage. No subscriptions, seat fees or key charges, and no hidden items.
| Token billing | Settled at per-million-token rates × actual usage; see the model pages — service fees already included |
|---|---|
| API Keys | Free to create, unlimited in number, each with an optional daily token quota |
| Request logs | Free logging of model, tokens, cost, latency, status and trace ID |
| Top-up channels | TRON (TRC20) · USDT |
| Minimum top-up | $1.00 |
| Subscription / seat fees | None |
Endpoints
Point your client's Base URL at either address below — the same key works for both; the model ID determines the price.
https://189.24.66.33/v1https://189.24.66.33How billing works
After each call, billing uses the real usage returned by the upstream: uncached input, cached input and output are priced separately. Amounts are recorded at high precision — short requests are never rounded up.
- Interrupted mid-stream:settled on upstream-confirmed usage; incomplete usage keeps the reservation pending review.
- Failed requests:requests that never entered generation release the reservation; started requests settle on confirmed usage.
- Insufficient balance:balance and quotas are checked before the upstream call; requests that cannot be covered are rejected outright (402).
- Per-key daily quota:each key can have its own daily token cap; once reached, requests are rejected for the day and resume the next day.
Model prices
Input, cached input and output are priced separately for each model; below are the current settlement rates — after price changes, the rate at call time applies.
| Model | Input / Mtok | Cached input / Mtok | Output / Mtok | Context |
|---|---|---|---|---|
| GLM 5.3glm-5.3-xploded | $2.00 | $1.00 | $5.00 | 1.05M |
| GLM 5.3 Flashglm-5.3-flash-xploded | $0.30 | $0.20 | $1.00 | 1.05M |
| Kimi K3.1 Prokimi-k3.1-pro-xploded | $7.50 | $0.80 | $37.50 | 1.05M |
| Tencent Hy4hy4-xploded | $3.20 | $0.60 | $6.40 | 1.05M |
| DeepSeek V4.1 Flash 1017deepseek-v4.1-flash-1017-xploded | $1.25 | $0.30 | $2.50 | 1.05M |
| DeepSeek V4.1 Flash Lite 1017deepseek-v4.1-flash-lite-1017-xploded | $0.36 | $0.03 | $1.44 | 512K |
| MiMo V2.6 Flashmimo-v2.6-flash-xploded | $1.00 | $0.50 | $2.00 | 1.05M |
| Qwen3.8 27Bqwen3.8-27b-xploded | $1.50 | $0.50 | $5.00 | 1.05M |
| Qwen3.8 27B Ultrafastqwen3.8-27b-fast-xploded | $1.75 | $0.50 | $5.50 | 524K |
Create an account and get a key
Top-up methods
Current channels: TRON (TRC20) · USDT. On-chain payments must transfer the exact amount to the network, asset, address shown on the order, and credit automatically after confirmation; manual channels are verified by an administrator against payment proof. Available channels, the minimum amount and order validity follow what the top-up page actually shows.
FAQ
Do I need a subscription?
No. Prepaid balance is pay-as-you-go, with no monthly minimum and no seat fees.
Are the listed prices what I actually pay?
Yes. Rates on the model pages and here include service fees and are used directly for settlement — what you read is what you pay.
Do extra API keys cost anything?
No charge. Create separate keys per client, project or script, and disable them any time.
What is the minimum top-up?
Currently $1.00. Available channels and payment instructions follow the order page; on-chain payments credit automatically after confirmation, manual channels are verified by an administrator.
What do request logs store?
Model, tokens, cost, latency, result and trace ID; never prompts or response text.
What if a request is interrupted or fails?
Settlement follows confirmed usage; reservations are released if generation never started. Interrupted requests with incomplete usage keep their reservation pending review, then settle or refund.
How is cached input priced?
Cache hits confirmed by the upstream are billed at the cached rate; regular input and cache writes are billed at the regular input rate.