LLM fallback cost calculator: worked examples
Route Lab compares short supplier sequences using entered cost, success and latency assumptions. It checks the worst-case total against your budget before sorting feasible options by expected cost. It is an offline planning tool; no model or payment is called.
Three fictional scenarios, calculated by the same engine as the tool. They illustrate behavior and failure cases, not customer outcomes or measured provider performance.
Cheap first, reliable second
Compare expected cost with the maximum you could spend.
- Feasible sequences
- 3
- Sequences checked
- 4
- Currency
- USD
| Sequence | Worst cost | Expected cost | Estimated success | Latency bound (ms) |
|---|---|---|---|---|
| Example A → Example B | 0.03 | 0.018 | 98.00% | 6000 |
| Example B | 0.02 | 0.02 | 95.00% | 4000 |
| Example B → Example A | 0.03 | 0.0205 | 98.00% | 6000 |
Run this scenario Input JSON Calculated report
Inspect every input
{
"currency": "USD",
"budget": "0.04",
"maxLatencyMs": 7000,
"minSuccess": 0.95,
"maxAttempts": 2,
"candidates": [
{
"id": "Example A",
"cost": "0.01",
"success": 0.6,
"latencyMs": 2000,
"allowed": true
},
{
"id": "Example B",
"cost": "0.02",
"success": 0.95,
"latencyMs": 4000,
"allowed": true
},
{
"id": "Restricted provider",
"cost": "0.005",
"success": 0.9,
"latencyMs": 1000,
"allowed": false
}
]
}Budget is too tight
Find out when none of the sequences meets every condition.
- Feasible sequences
- 0
- Sequences checked
- 4
- Currency
- USD
| Sequence | Worst cost | Expected cost | Estimated success | Latency bound (ms) |
|---|
Run this scenario Input JSON Calculated report
Inspect every input
{
"currency": "USD",
"budget": "0.015",
"maxLatencyMs": 7000,
"minSuccess": 0.95,
"maxAttempts": 2,
"candidates": [
{
"id": "Example A",
"cost": "0.01",
"success": 0.6,
"latencyMs": 2000,
"allowed": true
},
{
"id": "Example B",
"cost": "0.02",
"success": 0.95,
"latencyMs": 4000,
"allowed": true
},
{
"id": "Restricted provider",
"cost": "0.005",
"success": 0.9,
"latencyMs": 1000,
"allowed": false
}
]
}The fallback takes too long
A low expected cost cannot bypass a latency constraint.
- Feasible sequences
- 0
- Sequences checked
- 4
- Currency
- USD
| Sequence | Worst cost | Expected cost | Estimated success | Latency bound (ms) |
|---|
Run this scenario Input JSON Calculated report
Inspect every input
{
"currency": "USD",
"budget": "0.04",
"maxLatencyMs": 3000,
"minSuccess": 0.95,
"maxAttempts": 2,
"candidates": [
{
"id": "Example A",
"cost": "0.01",
"success": 0.6,
"latencyMs": 2000,
"allowed": true
},
{
"id": "Example B",
"cost": "0.02",
"success": 0.95,
"latencyMs": 4000,
"allowed": true
},
{
"id": "Restricted provider",
"cost": "0.005",
"success": 0.9,
"latencyMs": 1000,
"allowed": false
}
]
}A mistake worth catching
Two suppliers can fail on the same tasks. Multiplying independent failure rates can overstate reliability when they share models, infrastructure or task weaknesses.
Use the method with your records
Enumerate distinct allowed candidates up to three attempts. Reject sequences over total budget or latency, then compare expected cost among those meeting the supplied success target.
Offline planning only. No live requests, model judge, payment, dynamic bandit or production routing.
Read the complete method, sources and input contract. Updated 2026-09-19; by the AGI Scorecard team.