LLM Router overview

LLM Router gives you one API for 100+ LLMs from OpenAI, Anthropic, Google, and more. It automatically tries a fallback when a provider fails. You can route users or requests to different models, A/B test models and prompts, and measure the results: cost, latency, retention, and revenue.

New: video generation. The same API key now generates short video clips from a text prompt or images. See Video generation.

Your first request

Set INWORLD_API_KEY to the Base64 credential from the Portal, then send a chat completion. Use model: "auto" to let the router choose a model. To choose one yourself, set a provider/model ID such as openai/gpt-5.

curl --request POST \
  --url https://api.inworld.ai/v1/chat/completions \
  --header "Authorization: Basic $INWORLD_API_KEY" \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "auto",
    "messages": [
      {"role": "user", "content": "Write a 50-word bio for a software engineer."}
    ]
  }'

import os
from openai import OpenAI

api_key = os.environ["INWORLD_API_KEY"]
client = OpenAI(
    base_url="https://api.inworld.ai/v1",
    api_key=api_key,
    default_headers={"Authorization": f"Basic {api_key}"},
)

response = client.chat.completions.create(
    model="auto",
    messages=[{"role": "user", "content": "Write a 50-word bio for a software engineer."}],
)
print(response.choices[0].message.content)

The request follows OpenAI's chat completions format, so you can use the OpenAI SDKs without changes. The quickstart shows how to create a router in the Portal and send requests through it.

Explore

To generate and speak an answer in one API call, use Voice responses. To switch from another gateway, see Migrate from OpenRouter or Migrate from Anthropic.