GPT + 15 other models in one API

GPT-5 API
+ smart routing to save 90%

Use GPT-5 for complex tasks, auto-route simple queries to DeepSeek ($0.32/$0.48 per M tokens) or Qwen Turbo ($0.08/$0.30). Same API key, one line change.

Get API key — 10 trial calls

GPT-5

OpenAI latest flagship. 272K context, top benchmarks.

$2.88/ M input
$17.25/ M output

Model ID: openai/gpt-5

GPT-4o Mini

Fast everyday model. 128K context, great for simple tasks.

$0.17/ M input
$0.69/ M output

Model ID: openai/gpt-4o-mini

Same OpenAI SDK, more models

Change one line. Access GPT + Claude + DeepSeek + 13 more.

Python
from openai import OpenAI

# Just change the base_url — everything else stays the same
client = OpenAI(
    base_url="https://api.aipower.me/v1",  # ← only change
    api_key="YOUR_AIPOWER_KEY",
)

# GPT-5
r = client.chat.completions.create(
    model="openai/gpt-5",
    messages=[{"role": "user", "content": "Complex reasoning task..."}],
)

# Or save 91% with DeepSeek for simple tasks
r = client.chat.completions.create(
    model="deepseek/deepseek-chat",
    messages=[{"role": "user", "content": "Simple question"}],
)

# Or let AI pick the best model automatically
r = client.chat.completions.create(
    model="auto",  # Smart routing
    messages=[{"role": "user", "content": "Any task"}],
)

Get an API key → your API key auto-fills in all code examples. Copy & run instantly.

Why not just use OpenAI directly?

16 models, one key

GPT + Claude + DeepSeek + Qwen + 12 more. No managing separate accounts.

Smart routing saves 90%

Route simple tasks to $0.13/M Qwen Plus or $0.01/M GLM-4 Flash instead of $3.25/M GPT-5.

Auto-failover

If OpenAI goes down, requests auto-route to Claude or DeepSeek. 99.9% uptime.

WeChat Pay

Pay with credit card, WeChat Pay, or Alipay. No USD-only restriction.

Try GPT-5 + 15 other models

10 trial calls. No credit card. One line change from OpenAI.

Get API key