Cost model
The question is not “how much does Jev save” — it is how much quality you can afford to recover before the cost advantage disappears. The break-even fallback rate is where that line sits for your workload.
Your workload
These are placeholders describing a plausible classification workload. Replace them with your own measurements — nothing here is a finding.
The Jev cascade stays cheaper until 99.8% of requests fall back to Claude Haiku 4.5. Above that, running everything on the baseline costs less.
Based on your token usage and the verified provider pricing listed below. f* = (B − J) / F.
Assumptions
| Requests / month | 1,000,000 |
| Jev call | 300 in / 0 out |
| Baseline call | 500 in / 910 out |
| Fallback call | 500 in / 910 out → Claude Haiku 4.5 |
| Fallback rate | 0.0% |
| Batch pricing | not applied |
Pricing used
Observed escalation rates
These are measurements from specific runs, shown for scale. They are not defaults and none of them is “Jev’s typical fallback rate”.
Specific to this dataset, cascade and accuracy target. The run's pre-registered verdict was AMBIGUOUS: the difference against the comparison cascade was +0.265 with a 95% CI of [-0.530, +0.595].
The comparison cascade from the same run, under the same caveat.
The threshold was fitted on this same data, so the rate is optimistic by an unknown amount, and no single-model baseline was measured. The same pipeline found no workable threshold at all on a second dataset.
Can you route on Jev’s confidence?
Every fallback rate above assumes you can tell in advance which requests Jev will get wrong. The recorded evidence splits cleanly, and it splits by how hard the judgment is — not by anything you can read off a spec sheet.
Topic membership on 20 Newsgroups. All six pre-registered gates passed.
Graded product relevance on Amazon ESCI. Same author, same gates, same model version — four of six failed.
Agent tool-call risk. Every incorrect answer came with hedged confidence; the model never returned 1.000 and was wrong.
Error-ranking AUROC from Banking77 to CLINC150. On CLINC150 Jev ranks its own errors worse than the nano-class baseline does.
A routing pipeline found a workable threshold on Banking77, escalating 11.6% of rows — though it was fitted on that same data.
The same pipeline on Web of Science: no threshold beat the better single model.
How this is calculated
Three scenarios are priced for the same workload:
- B — baseline only: one call per request to the model you use today.
- J — Jev only: one Jev call per request.
- H(f) — hybrid cascade: one Jev call per request, plus one fallback call on a fraction f of requests. H(f) = J + f·F.
Setting H(f) = B and solving gives the break-even fallback rate f* = (B − J) / F. Below it the cascade is cheaper; above it the baseline is.
Four decisions that change the answer
Output tokens are priced on both sides. Jev bills output at $0.00/MTok; a generative baseline does not, and output typically costs several times input. A calculator that asks only for “average input tokens” understates the baseline — usually by an order of magnitude on this shape of workload — and inflates the apparent saving.
Jev’s input tokens are independent of the baseline’s. A typed decision request and a generative prompt with a system prompt, schema and examples are not the same length.
The fallback call is modelled separately. F is not assumed equal to B — an escalation call may carry Jev’s output, a different prompt, or go to a different model. When F does equal B, the formula reduces to f* = 1 − J/B; we let that fall out of the arithmetic rather than baking it in.
Cache and batch rates are provider-specific. We never apply a house discount. Ask for a cached-input share on a model with no published cache price and the model returns an error instead of a number, because that one assumption moves the result more than anything else on the page — in the direction that flatters Jev.
Full reasoning and the pricing record format are in the methodology.