AI model routing API for developers

Who it is for

Developers comparing enabled model routes for chat, responses, coding, agents, long context, cost control, and fallback behavior.

Use cases

• Model evaluation.
• Fallback planning for enabled routes.
• Cost control for repeated workloads.
• Agent routing experiments.
• Long-context workflow testing.

How InferGate connects

Create an API key, call /v1/models to inspect enabled models, then send compatible requests to selected model names.

Base URL

https://api.useinfergate.com/v1

curl example

curl https://api.useinfergate.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_INFERGATE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [
      {"role": "user", "content": "Evaluate this model route."}
    ]
  }'

JavaScript example

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.INFERGATE_API_KEY,
  baseURL: "https://api.useinfergate.com/v1"
});

const completion = await client.chat.completions.create({
  model: process.env.INFERGATE_MODEL || "gpt-4o-mini",
  messages: [{ role: "user", content: "Evaluate this model route." }]
});

console.log(completion.choices[0]?.message?.content);

Python example

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.environ["INFERGATE_API_KEY"],
    base_url="https://api.useinfergate.com/v1",
)

completion = client.chat.completions.create(
    model=os.getenv("INFERGATE_MODEL", "gpt-4o-mini"),
    messages=[{"role": "user", "content": "Evaluate this model route."}],
)

print(completion.choices[0].message.content)

Trial credits and pricing

New accounts currently include $20 in trial credits. Pricing depends on selected model/provider route and dashboard configuration. Use prepaid API credits when ready to scale.

FAQ

Can I route every model?

No. You can route enabled models available to your account.

How do I check models?

Use the dashboard or call /v1/models with your API key.

Does routing change pricing?

Pricing can depend on selected model/provider route and current account configuration.

Can I use routing for fallback?

Use enabled routes to design fallback behavior, then test your exact workload before production.