Create an account first, complete a small top-up or use a redemption code if available, then confirm your balance before your first real request.
If you are a light personal user, start with the $5 trial first. The $50 path is for people who already know they need a larger monthly envelope.
If you want something a little bigger than a trial but still small, the $10 personal path is the default fit.
Credits are used as API balance. Text models are billed by input and output tokens; media workflows are quoted by configuration. Register first so manual PayPal top-ups or redemption codes can be attached to the right account.
After creating a Black Eagle AI account, pay through PayPal and send your account email plus payment receipt in the same channel where you contacted Black Eagle AI, or email lianzhiweivincent@gmail.com. Credits are added after confirmation so the first buyer can verify one real route before topping up more. If support sends a redemption code, you can use it for faster crediting. If you are coming from a Qwen or LiteLLM issue, start at request access.
Read the billing FAQ before topping up if you want the exact step-by-step flow.
Customer-facing rates are listed per 1M tokens. Actual usage is metered separately for input and output tokens.
| Model tier | Example models | Input | Output | Access |
|---|---|---|---|---|
| Economy TextFast, low-cost chat and app workflows. | deepseek-v4-flash, qwen-flash | $0.40 / 1M tokens | $0.90 / 1M tokens | Available |
| Reasoning TextReasoning, coding and more complex answers. | deepseek-v4-pro | $1.00 / 1M tokens | $2.20 / 1M tokens | Available |
| Standard Chinese TextGeneral-purpose Chinese model families. | qwen-plus, GLM-4.7 | $1.20 / 1M tokens | $3.00 / 1M tokens | By request |
| Premium TextStronger models and higher-complexity workloads. | kimi-k2.5, glm-5, qwen3-max | $2.00 / 1M tokens | $6.00 / 1M tokens | By request |
| Media WorkflowsVideo and creative generation jobs. | Kling, Seedance, other media models | Custom quote by model, duration, resolution and volume | Early access | |
Model names and availability may change as providers update APIs, rate limits and regions. DeepSeek's legacy aliases deepseek-chat and deepseek-reasoner are scheduled for retirement on July 24, 2026, so new workloads should start on deepseek-v4-flash or deepseek-v4-pro. Browse the model catalog and contact support before moving production traffic.
These examples assume equal input and output token volume. Real usage depends on prompt size, output length and model choice.
About 3.8M input + 3.8M output tokens on Economy Text.
About 15.3M input + 15.3M output tokens on Economy Text, or 6.2M + 6.2M on Reasoning Text.
About 38.4M input + 38.4M output tokens on Economy Text, or 6.2M + 6.2M on Premium Text.
Billing is based on metered token usage and account balance. Model availability, supported features and rate limits may vary by provider. Start by registering and running one request, then use the manual PayPal top-up link or a redemption code after support confirms your account. For the exact trust and payment flow, read the FAQ.