Model Cost Breakdown

Paste trace exports to see model share, request volume, and average cost. Copy JSON, Download CSV, or Reset to start a new trace.

Validate the model mix behind the bill

Use this view when spend drifts and you need to confirm whether a rate change, a request spike, or one model line moved first.

  1. 1Paste the trace window you want to validate.
  2. 2Check whether one model now owns most of the spend.
  3. 3Export the breakdown once the rate drift is clear.

Results update live as you paste or type.

Live Preview

Live preview

Add valid input to populate this result panel.

Paste model-level trace JSON or NDJSON above to see spend share, request volume, and concentration flags.

Why this tool

Find the model line that moved the bill

Use the free model breakdown when one invoice line jumps and you need to confirm whether the change came from pricing, request volume, or routing.

Spot whether one model now owns most of the bill before the invoice review starts.

Check if a rate change, request spike, or routing choice caused the jump in spend.

Share a clean breakdown with finance or engineering without rebuilding the data by hand.

Works with traces from

  • OpenAI
  • Anthropic Claude
  • Google Gemini
  • Mistral
  • Cohere
  • AI21

How it works

Go from raw trace records to a clean model summary

The tool keeps the free check interactive, then gives you the next route to open when the question moves from model choice to owner or time-window comparison.

  1. 1

    Paste a trace window

    Drop in JSON or NDJSON records from provider, gateway, or app logs so the tool can read model names, requests, and cost.

  2. 2

    Review the model split

    See cost share, request volume, and average cost per request in one table to confirm what moved first.

  3. 3

    Carry the answer forward

    Export the breakdown, compare it against another window, or move into owner-level attribution when the next question changes.

Benefits

Keep the next answer moving

Compare two windows

Open a before-and-after diff when a deploy or provider change moved the bill and you need the delta in one table.

Open Compare

Group spend by owner

Switch to attribution when finance needs the same spike tied back to a team, project, or customer.

Open Attribution

Saved reports

Save the monthly model review

Join the waitlist if the same model-spend check needs to happen on a schedule or needs to be shared with the team.

See pricing

Need access to an existing saved workspace? Sign in.

FAQ

Questions teams ask before they save the report

Keep using the free check for one-off reviews, then move to saved reports when the same model-cost question comes back each month.

Why this tool

The pricing page says GPT-4 is $30/1M tokens, but your invoice says otherwise — because some tenants are on a different tier and some rates changed mid-month. Model Pricing rolls your trace into per-model cost, request, and average-cost-per-request so you can spot the model line where your assumed rate and your real rate diverge.

How it works

  1. 1

    Export the trace JSON for the period you want to validate.

  2. 2

    Paste it above; the model table updates live as you type.

  3. 3

    Compare avg-cost-per-request against your assumed rate; investigate any model flagged as a concentration anomaly.

Benefits

Per-model cost, share, and average cost per request in one table.
Flags any model holding ≥ 50% of spend so concentration risk is obvious.
Reveals rate drift — if avg-cost-per-request shifts, the assumed rate is wrong.
No vendor login, no sign-up — paste and read.

Frequently asked

Catch rate drift the day it happens

Pro saves per-model history and alerts when avg-cost-per-request shifts.

Create free account →

Like what you see? Get early access to the full Auditor

Join the waitlist to save audit sessions and turn trace data into attribution reports.

Join waitlist