A couple weeks ago I wrote about why I chose New API as my gateway backend. Since then the most common question has been: "ok, how do I actually use it?" So here's the 5-minute setup.
If you're managing multiple API keys for different AI models, you know the pain. One key for OpenAI, another for Claude, a third for DeepSeek. Different billing dashboards, different SDKs, different endpoints.
AIOpenCloud wraps them all behind one endpoint. Here's how to start using it in 5 minutes.
Step 1: Get your key
Sign up at aiopencloud.xyz. You get $8.88 free credit — no credit card needed.
Step 2: Install the client (optional)
You don't actually need a client. Any OpenAI-compatible SDK works. But for a quick test:
pip install openai
Step 3: Make your first call
from openai import OpenAI
client = OpenAI(
base_url="https://aiopencloud.xyz/v1",
api_key="your-key-here"
)
response = client.chat.completions.create(
model="deepseek-v4-flash",
messages=[{"role": "user", "content": "Hello, what models do you support?"}]
)
print(response.choices[0].message.content)
That's it. One endpoint. One key. All models.
What you get:
- DeepSeek, GPT, Claude, Gemini, Qwen — all through the same API
- One dashboard for usage and billing
- No lock-in — your key works with any OpenAI SDK
Pricing example
| Model | Input (per 1M tokens) |
|---|---|
| DeepSeek V4 Flash | $0.16 |
| GPT-5 | $6.00 |
| Claude Sonnet 4 | $3.60 |
| Gemini Flash | $0.60 |
Full pricing comparison coming soon at aiopencloud.xyz/pricing-vs
What's your setup for managing multiple AI models? Still juggling separate keys, or found a workflow that works? I'd love to hear what others are doing — drop a comment.
Next up: a breakdown of what models people actually use vs what they think they need. Follow if you want the real numbers.
Top comments (0)