Skip to content
AZRouter.dev

Get started

Quickstart

Make your first AZRouter request in curl, Python or TypeScript.

If your code already calls OpenAI or Anthropic, the switch is two values: the base URL and the key.

1. Add credits and create a key

New accounts start on the Free tier, whose monthly cap is $0. Add credits in Dashboard → Credits first, then create a key in Dashboard → API keys. The key is shown once; copy it and export it so the examples below can read it.

Shell
export AZROUTER_API_KEY="sk-..."

2. Pick a base URL

Base URL: https://api.azrouter.dev/v1

Use it for the OpenAI SDKs and any tool with an "OpenAI-compatible" or "OpenAI base URL" setting. The client adds /chat/completions, /responses, /embeddings and so on.

Shell
export OPENAI_BASE_URL="https://api.azrouter.dev/v1"
export OPENAI_API_KEY="$AZROUTER_API_KEY"

3. Send a request

The examples use qwen/qwen3.8-flash, a low-cost text model. Any slug from the model list works the same way.

curl https://api.azrouter.dev/v1/chat/completions \
  -H "Authorization: Bearer $AZROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen/qwen3.8-flash",
    "messages": [{"role": "user", "content": "Reply with exactly: Hello from AZRouter"}],
    "max_tokens": 200
  }'

The reply text is in choices[0].message.content:

Output
Hello from AZRouter

Leave room for reasoning

Some models, qwen/qwen3.8-flash among them, think before they answer and bill those reasoning tokens as output. With a very small max_tokens the limit can run out during the thinking, and content comes back null with finish_reason: "length". Keep max_tokens at a few hundred for short answers.

Next steps