DEV Community

chenxiao5580-cmd
chenxiao5580-cmd

Posted on

Call GPT, Claude, and Gemini from one API key — a 3-step setup

#beginners #tutorial #ai #api

If you want to try GPT, Claude, and Gemini without signing up for three separate platforms and juggling three billing dashboards, here's a 3-step setup using an OpenAI-compatible gateway.

1. Create an account

Go to modelishub.com, sign up with a username/email + password, and pass the quick human check. You're logged into the console immediately — no approval wait.

2. Create an API key

In the console, open API Tokens → New Token and give it a name (e.g. my-app). Optionally set a quota cap so you can control spend per project. Copy the key (it starts with sk-) — that's what you call the API with. Keep it private; usage and quota are tracked per key.

3. Make your first call

Point any OpenAI-compatible tool at https://modelishub.com/v1 with that key:

from openai import OpenAI

client = OpenAI(base_url="https://modelishub.com/v1", api_key="YOUR_KEY")
resp = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)
Enter fullscreen mode Exit fullscreen mode

To switch models, just change the model id — claude-sonnet-4-6, gemini-2.5-flash, and so on. Same endpoint, same key. Fetch the full list any time with GET /v1/models.

Checking your usage

The console Dashboard shows overall consumption; the Logs view shows every call (model, tokens used, quota spent, success or error), so you can pinpoint a wrong model id or a quota issue in seconds.

This runs on Modelis, an OpenAI-compatible LLM gateway that auto-routes each request to a fitted model and bills a flat per-call price (not per-token). Free tier to try it — get a key.

Top comments (0)