Illustrative 70B inference node-rental cost at stipulated $19.20/hour and continuous 2,800 output tokens/second
$1.90/M output tokens (illustrative node rental)derived
| Value kind | derived — Derived values are guide calculations; their result is only as fixed as the stated inputs, scope, and method. |
|---|---|
| Scope | Stipulated node-rental illustration, not current market pricing or complete serving COGS; excludes idle time, retries and non-node costs. |
| As of | 2026-08-24 |
| Source | Guide derivation from the illustrative Chapter 1.8 inputs · Chapter 1.8, inference unit-economics example |
| Derivation | 19.20 x 1,000,000 / (3,600 x 2,800) = 1.9047619 dollars per million generated output tokens, assuming continuous stated throughput |
| Review | checking…review by 2027-04-05 · standard cadence |
| Recorded changes | last 2026-10-05 · 5 revisions tracked |
| Claim id | inference-cost-per-million-tokens-self-hosted |
Where the guide uses it
← Full numbers register — every date-stamped figure in the guide, with revision history.