Know what every AI call costs.

Per-request cost across every provider. See where your tokens go, which models cost the most and where to optimise.

app.grepture.com/dashboard
Today's spend
$12.47
Requests today
1,847
Avg cost/req
$0.0067
Top model
gpt-4o
ModelRequestsCostShare
gpt-4o892$8.9471.7%
claude-sonnet-4412$2.5520.4%
gpt-4o-mini389$0.544.3%
gemini-2.5-pro154$0.443.5%

What it does

Grepture counts tokens and calculates cost for every request your app makes, broken down by request, model and endpoint, across OpenAI, Anthropic, Google AI and every other supported provider. It works in proxy mode and trace mode.

How it works

Input and output tokens are counted per request and mapped to each provider's pricing. View spend in the dashboard, filter by time range, model or endpoint, and export it for billing. In trace mode the SDK reads token counts client-side and sends them in the background: the same cost data with no proxy hop.

Know what every request cost.

  • Per request

    Input and output tokens, priced at each provider's rates, on every request.

  • Per model and endpoint

    See which model, provider or endpoint drives spend, and how it trends over time.

  • Ready for chargeback

    Export spend for internal billing by team or feature.

  • No proxy required

    In trace mode the SDK reads token counts client-side, so you get the same cost data without the proxy hop.

What's included

  • Token count and cost for every request
  • Cost breakdown by model, provider and endpoint
  • Spend trends and usage over time
  • Pricing for all major providers
  • Filter and search by cost, tokens or model
  • Export for internal billing and chargeback
  • Trace mode: cost tracking without the proxy hop

Send your first clean request in five minutes.

Free for 1,000 requests a month. No credit card, no sales call.