Quickstart

Inferel’s API is OpenAI-compatible: any OpenAI SDK works by changing the base URL and key.

Base URL


OpenAI-compatible clients	`https://inference.inferel.ai/v1`
Anthropic-compatible clients	`https://inference.inferel.ai`
Console (keys, billing, usage)	console.inferel.ai

1. Create an account and API key

Sign up or sign in at console.inferel.ai.
Go to API Keys and click Create key.
Copy the key immediately — for security it is shown only once. Store it in an environment variable:

export INFEREL_API_KEY="your-key-here"

New accounts need a credit balance before making paid requests — see Billing & credits to top up.

2. Make your first request

cURL

curl https://inference.inferel.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $INFEREL_API_KEY" \
  -d '{
    "model": "claude-sonnet-4.6",
    "messages": [
      {"role": "user", "content": "Say hello from Inferel!"}
    ]
  }'

Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://inference.inferel.ai/v1",
    api_key="YOUR_INFEREL_API_KEY",
)

response = client.chat.completions.create(
    model="claude-sonnet-4.6",
    messages=[{"role": "user", "content": "Say hello from Inferel!"}],
)
print(response.choices[0].message.content)

Node.js (OpenAI SDK)

import OpenAI from 'openai';

const client = new OpenAI({
  baseURL: 'https://inference.inferel.ai/v1',
  apiKey: process.env.INFEREL_API_KEY,
});

const response = await client.chat.completions.create({
  model: 'claude-sonnet-4.6',
  messages: [{ role: 'user', content: 'Say hello from Inferel!' }],
});
console.log(response.choices[0].message.content);

3. Stream the response

Set stream: true to receive tokens as server-sent events. Add stream_options={"include_usage": True} when you also want the final usage chunk:

stream = client.chat.completions.create(
    model="claude-sonnet-4.6",
    messages=[{"role": "user", "content": "Write a haiku about GPUs."}],
    stream=True,
    stream_options={"include_usage": True},
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

Next steps

Pick a model from the live catalog — GET /v1/models always reflects what’s currently available, with each model’s per-token price.
Understand rate limits and how to set per-key limits.
Learn how errors are reported so your retries behave well.