Loading page…
Every model on the gateway, billed per token or per generated item from one prepaid balance. Live rates come from the connected New API pricing service.
| Model | Price | Endpoints |
|---|---|---|
| deepseek-v4-flash | — in— out | openai, openai-response, openai-response-compact … |
| deepseek-v4-pro | — in— out | openai, openai-response, openai-response-compact … |
| Model | Price | Endpoints |
|---|---|---|
| MiniMax-H3 | — in— out | openai |
| Model | Price | Endpoints |
|---|---|---|
| kimi-k3 | $3.00 in | openai, openai-response, openai-response-compact … |
| Model | Price | Endpoints |
|---|---|---|
| gpt-5.5 | $5.00 in | openai |
| gpt-5.6-sol | $4.00 in | openai |
| gpt-5.6-terra | $2.00 in | openai |
| gpt-6-astra | $10.00 in$50.00 outcached $1.00 /1M | openai |
| gpt-image-2 | $8.00 in$30.00 outcached $2.00 /1M | image-generation, openai |
| Model | Price | Endpoints |
|---|---|---|
| grok-4.5 | — in— out | openai, openai-response |
| grok-4.6 | — in— out | openai, openai-response |
| grok-4.7 | — in— out | openai, openai-response |
| grok-imagine-image | — in— out | openai, openai-response |
| grok-imagine-image-2.0 | — in— out | openai, openai-response |
| grok-imagine-image-quality | — in— out | openai, openai-response |
| hy4-preview | $0.834 in$2.50 outcached $0.042 /1M | openai |
| Model | Price | Endpoints |
|---|---|---|
| doubao-seedance-2-0-260128 | — in— out | openai |
| doubao-seedance-2-0-fast-260128 | — in— out | openai |
| doubao-seedance-2-0-mini-260615 | — in— out | openai, seedance |
| doubao-seedance-2-5-260628 | — in— out | openai |
| Model | Price | Endpoints |
|---|---|---|
| grok-imagine-video-1.5-preview | $75.00 in$75.00 out | openai, openai-response |
| Model | Price | Endpoints |
|---|---|---|
| glm-5.2 | $1.40 in$4.40 outcached $0.260 /1M | openai |
| glm-5.3 | $1.40 in$4.40 outcached $0.260 /1M | openai, openai-response, openai-response-compact … |
| glm-5.3-flash | $0.150 in$0.500 outcached $0.030 /1M | openai, openai-response, openai-response-compact … |
Token-based models: USD / 1M tokens = model_ratio × 2 × group_ratio — output tokens apply the model's completion ratio, and cached input (where offered) is billed at the cache ratio. Per-item models (most image and video generation) bill a fixed price per generated item.
This page shows the default group rates returned to anonymous visitors (known groups: 10%off, 20%off, 30%off, 40%off, 5%off, minimax). Your account group may carry a different multiplier — sign in and open the console pricing page to see the exact rates applied to your key.
Any comparison price is shown only when it has been verified for the current model and pricing snapshot. The live New API catalog remains the source of truth; this page refreshes hourly.
pricing_version: a42d372ccf0b5dd13ecf71203521f9d2
Chat models are billed per token: input, output, and (where supported) cached input are metered separately. Image and video models are billed per generated item. Everything deducts from one prepaid balance shared across every model.
The prices shown are the default-group rates an anonymous visitor sees. After signing in, your account group may carry a different multiplier — the console shows the exact rates applied to your key.
EasyAI uses a prepaid balance. There is no minimum top-up, the balance does not expire, and usage is deducted as requests use enabled models. Payment processing is handled through the configured payment provider.
Billing follows the usage record shown by the gateway. If you see a duplicate or incorrect charge, contact support with the request ID and transaction details for review.
When a provider changes its price or availability, the live catalog may change. Check the current pricing and model details before production use.
Yes. A single key routes to every model your account group can access — switching provider is a one-line model name change in your request.