SubscriptionsDeepSeek Access
DeepSeek Access

DeepSeek Access

API Key Product

Get access to DeepSeek V3 and DeepSeek R1 — powerful models for coding, reasoning and general AI tasks at an affordable price. Use any OpenAI-compatible client.

What's included

  • DeepSeek V3 & R1 models
  • Single API key
  • Fast activation
  • Usage support

How to use your DeepSeek API key

1

Get your API key

After purchase, your dedicated API key will be sent via WhatsApp and available in your dashboard under API Keys.
2

Install Chatbox on any device

Download **Chatbox** — a free app for all platforms:

- **Windows**: Download for Windows
- **macOS**: Download for Mac
- **Linux**: Download for Linux
- **Android**: Get it on Google Play
- **iOS**: Get it on the App Store
- **Browser**: Use the web version directly — no install needed

It works seamlessly with custom OpenAI-compatible APIs.
3

Configure in Chatbox

Open Chatbox → Settings → **AI Provider** → **OpenAI API Compatible**. Set:

- **API Key**: Your DeepSeek key
- **API Domain**: `https://api.virtusub.xyz/v1`

Click **Save**.
4

Pick a model

Select **deepseek-chat** (DeepSeek V3) for general use, or **deepseek-reasoner** (DeepSeek R1) for complex reasoning with chain-of-thought.
5

Use with API (advanced)

Your key is OpenAI-compatible. Use it directly with the OpenAI Python SDK:

```python
from openai import OpenAI
client = OpenAI(api_key="your-key", base_url="https://api.virtusub.xyz")
response = client.chat.completions.create(model="deepseek-chat", messages=[{"role":"user","content":"Hello"}])
```

Pro tip

All our AI models use a single compatible API format. If you can configure one, you can configure them all. The base URL https://api.virtusub.xyz/v1 works for every AI model key.

DeepSeek Access

AI Models

DeepSeek Access

From ₦10,000/mo

One API key for DeepSeek models

Frequently asked questions

What's the difference between V3 and R1?

DeepSeek V3 is fast and great for everyday tasks. DeepSeek R1 uses chain-of-thought reasoning — it 'thinks' before answering, making it better for math, logic and complex problems.