At what scale does self-hosting an LLM pay off?
Planning ahead? To find the scale at which self-hosting an LLM pays off, you need to see how today's per-user cost behaves as volume climbs. This page helps teams planning for scale project the bill at scale before the growth (and the invoice) actually arrives.
Results update automatically as you type.
- API cost / request
- $0.000645
- GPU host / month
- $1,500.00
- Break-even volume
- 2,325,581 req/mo
Why this isn't trivial
The part people underestimate: self-hosting's fixed GPU cost only beats per-token pricing above a break-even volume, which many teams reach later than they expect. In practice the biggest savings come from staying on the API until volume clears the break-even, so it is worth modelling before you commit.
How it's calculated
We estimate this by finding the request volume where amortized GPU cost per request equals the API price. Every figure uses the current provider prices baked into the site (reviewed daily), and you can override any input to match your own assumptions.
Frequently asked questions
What's the break-even volume?+
Where fixed GPU cost spread over requests drops below the API's per-request price.
Should I switch as soon as I hit it?+
Factor ops burden too; clear the break-even with margin before switching.
Are these prices up to date?+
Yes. The model prices behind this calculator are refreshed and reviewed daily, so your estimate reflects current provider rates rather than a stale snapshot.
Related
Estimates for planning. Pricing data last reviewed 28 July 2026.