~5 minutes to complete

Get Started with AImonitor

Start tracking and optimizing your AI costs in 6 simple steps.

Prerequisites

  • An AImonitor account (free to create)
  • API key from at least one AI provider (OpenAI, Anthropic, etc.)
  • Your application code ready to update
1

Create Your Account

Sign up for a free AImonitor account. No credit card required.

Create Free Account

Already have an account? Log In

2

Create a Project

Projects help you organize API usage by application or environment (e.g., "Production", "Staging", "Development").

  1. 1Go to your Dashboard
  2. 2Click "New Project" in the sidebar
  3. 3Enter a project name (e.g., "Production API", "My SaaS App")
  4. 4Click "Create"
3

Get Your AImonitor API Key

Your AImonitor API key authenticates requests through our proxy. This is different from your provider keys.

  1. 1Open your project settings (click the gear icon)
  2. 2Go to the "API Keys" tab
  3. 3Click "Create API Key"
  4. 4Give it a name (e.g., "Production Server")
  5. 5Copy and save your key immediately
Your key will look like: aim_sk_1a2b3c4d5e6f7g8h9i0j...

Important

Save your API key immediately after creation. For security reasons, you won't be able to see it again. If you lose it, you'll need to create a new one.

4

Add Your Provider API Keys

Add the API keys for each AI provider you use. We encrypt them at rest and never log them.

  1. 1In project settings, go to "Provider Keys"
  2. 2Click on the provider you want to add
  3. 3Paste your API key from that provider
  4. 4Click "Save"

OpenAI

sk-...

Anthropic

sk-ant-...

Google AI

AIza...

Mistral

...

Cohere

...

Azure

...

5

Update Your Code

Change the base URL in your existing code. Everything else stays exactly the same.

OpenAIPython
from openai import OpenAI

client = OpenAI(
    base_url="https://proxy.aimonitor.ai/openai/v1",
    default_headers={"X-AIMonitor-Key": "aim_your_key_here"}
)

# Your existing code stays exactly the same
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello, world!"}]
)

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

Key changes:

  • base_url → points to AImonitor proxy
  • X-AIMonitor-Key → your AImonitor API key
  • • Everything else (model, messages, parameters) stays exactly the same
6

Verify Your Setup

Make a test API call and verify it appears in your dashboard.

Run your application

Make at least one API call using your updated code

Check your dashboard

Go to your AImonitor dashboard — the request should appear within 5-10 seconds

Verify the details

Check that the request shows the correct model, token count, and calculated cost

Troubleshooting

401 Unauthorized error

Check that your X-AIMonitor-Key header is set correctly. Verify the API key in your project settings. Make sure there are no extra spaces or characters.

Requests not showing in dashboard

Ensure you're using the correct project API key. Logs typically appear within 5-10 seconds. Try refreshing the dashboard.

Provider key not found error

Add your provider's API key (e.g., OpenAI sk-... key) in your project settings under "Provider Keys". Each provider needs its own key.

Connection timeout

Our proxy adds <50ms latency. If you're seeing timeouts, check your network connection. For very long requests, increase your client timeout.

Invalid API key format

AImonitor keys start with "aim_". Provider keys have their own formats (OpenAI: sk-..., Anthropic: sk-ant-...). Don't mix them up.

CORS errors in browser

AI API calls should be made from your backend, not directly from the browser. This protects your API keys and avoids CORS issues.

Still having issues? Contact support or check our full documentation.

Next Steps

DEVELOPMENT