Gross Margin Per Request Calculator

Check margin when you charge per AI request.

Quick answer

With the default inputs, $0.0994 margin per request — 99.4% gross margin. Enter your own numbers below to recompute instantly; the full step-by-step math is shown under the worked example.

Model

Revenue per request($)

Input tokens / request(tokens)

Output tokens / request(tokens)

Results update automatically as you type.

Result

$0.0994 margin per request

99.4% gross margin

Revenue / request: $0.1000
AI cost / request: $0.000585
Margin / request: $0.0994
Gross margin: 99.4%

Enter what you charge per request and the workload behind it to see your AI cost, margin and gross-margin percentage — and a warning if you're underwater.

How this is calculated

AI cost per request = input tokens × input price + output tokens × output price, at your model's live rates. Gross margin = revenue per request − AI cost per request. Gross-margin % = margin ÷ revenue. If AI cost exceeds revenue, the margin is negative and you lose money on every call — the calculator flags this.

Is this a good result? What to do next

Software businesses typically target 70%+ gross margin, and AI token cost eats directly into that. A per-request margin above ~70% is healthy; between 40–70% is workable but watch it as volume grows; below that, or negative, you need to act. Remember this is model cost only — real gross margin is lower once support and infra are included.

Typical planning ranges

Healthy gross margin: ~70%+
Watch closely: ~40–70%
Act now: < 40% or negative

Ranges are typical planning figures to sanity-check your result, not authoritative benchmarks. Your numbers will vary with use case, volume, and vendor.

How to improve this number

Raise price per request or move to value-based pricing.
Use a cheaper model or shorten prompts and cap output tokens.
Cache repeated context so input tokens aren't re-billed.

Common mistakes

Treating model cost as total cost — infra and support reduce true margin.
Optimising the average while a few heavy requests run at a loss.
Forgetting output tokens cost more than input.

When to use a different approach

To set a resale price from a target markup, use the AI API markup calculator. To view economics per user rather than per request, use the AI cost per user calculator.