The price of every request, in the response

Cost Breakdown

Get real-time cost information for each request directly in the response's usage object.

Get startedRead the docs
response.usage
"usage": {
  "prompt_tokens": 1240,
  "completion_tokens": 312,
  "cost": 0.00467,
  "cost_details": {
    "input_cost": 0.0031,
    "output_cost": 0.00157,
    "total_cost": 0.00467
  }
}
Real cost, per request
01

In every response

No extra API call.

02

Itemized

Input, output and cache costs.

03

Streaming too

Cost arrives with the last chunk.

04

Bill your users

Exact cost per end user.

Built for

  • Usage billing

    Charge customers what they use

  • Budget guards

    Stop runaway agents in code

  • Feature costing

    Know what each feature costs

Read the cost
app.ts
1const completion = await client.chat.completions.create({2	model: "gpt-4o",3	messages: [{ role: "user", content: "Hello!" }],4});5 6console.log(completion.usage?.cost);
Read docs

More in observability & cost

All features

Request Metadata

Tag requests, filter later

Source Attribution

Get credit for your traffic

Data Retention

Keep metadata, or keep everything

Start using Cost Breakdown today

One API key. Every model. Live in minutes.

Start buildingBrowse models