Skip to main content

Quick Start

Get your first AI API call routed through AI SpendOps in under 5 minutes.

Prerequisites​

  • An AI SpendOps account with an API key (aso_k_...)
  • An API key for your AI provider (e.g. OpenAI sk-...)

Step 1: Understand the URL format​

The proxy URL follows this pattern:

https://proxy.aispendops.com/v1/{provider}/{provider-api-path}

Everything after /v1/{provider} is forwarded to the provider's upstream API as-is.

Step 2: Make your first request​

Add two things to your existing API call:

  1. Change the base URL to https://proxy.aispendops.com/v1/{provider}
  2. Add the X-ASO-API-Key header
curl https://proxy.aispendops.com/v1/openai/v1/chat/completions \
  -H "Authorization: Bearer sk-your-openai-key" \
  -H "X-ASO-API-Key: aso_k_yourkey.secret" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4.1",
    "messages": [{"role": "user", "content": "Hello, world!"}]
  }'

You'll get back the exact same response as calling OpenAI directly, plus an X-ASO-Request-Id header for tracing.

Step 3: Add dimensions (optional)​

Tag your requests with metadata for cost attribution:

curl https://proxy.aispendops.com/v1/openai/v1/chat/completions \
  -H "Authorization: Bearer sk-your-openai-key" \
  -H "X-ASO-API-Key: aso_k_yourkey.secret" \
  -H "X-ASO-Dims: team=backend,app=chatbot,env=prod" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4.1",
    "messages": [{"role": "user", "content": "Hello, world!"}]
  }'

Step 4: Use with your SDK​

Most AI SDKs let you override the base URL. Here's the Python OpenAI SDK:

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-openai-key",
    base_url="https://proxy.aispendops.com/v1/openai/v1",
    default_headers={
        "X-ASO-API-Key": "aso_k_yourkey.secret",
        "X-ASO-Dims": "team=backend,app=chatbot",
    },
)

response = client.chat.completions.create(
    model="gpt-4.1",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

See SDK Integration for more examples.

What happens behind the scenes​

  1. The proxy authenticates your X-ASO-API-Key
  2. Your request is forwarded to the provider unchanged
  3. The provider's response is returned to you unmodified
  4. Asynchronously (after response delivery), usage data is extracted and queued for processing
  5. Usage events are enriched with pricing and written to the analytics database

The async extraction means zero overhead on your response latency.

Next steps​