Cost formula
Blended monthly cost = (primary uptime × primary cost) + (failover rate × secondary cost); Failover premium = blended cost − primary-only cost
Calculator
Multi-provider failover cost calculator
Model the cost of an availability path: try the primary, and only on failure call a secondary provider. Compare ideal primary spend, failover overhead, and an all-secondary baseline.
Failover bill
Blended cost with failover
$497.60+$197.60 vs ideal
99.84% jobs complete · 16,000 hit Anthropic
Primary spend$284.00
Failover spend$213.60
Ideal primary$300.00
All on failover$2,700.00
Cost / job$0.002488
Failover share8.0%
Path
Primary → failover
Elevated errors; primary often bills input on timeout
Blend
Primary vs failover bill
Dual scales show how much of the window lands on each provider.
Failover job share8.0%
Lower primary success shifts spend to Anthropic. Current primary success 92%.
Primary successes184,000
Failover successes15,680
Abandoned320
Overhead vs ideal$197.60 (65.9%)
Availability failover is not the same as quality cascade routing—here the secondary runs only when the primary fails. Updated 2026-07-31. Approximate static list prices for planning only. Always verify against each provider’s official pricing page before production budgeting.
Guide
How to get value from this calculator
Estimate spend when primary provider failures route traffic to a secondary provider with different pricing. Failover improves reliability but secondary models may cost more per token or lack cache parity. Budget the failover premium instead of assuming primary pricing holds during outages.
Why it matters
Multi-provider resilience is an insurance policy with a premium. During outages you pay secondary list prices and may lose cache benefits—modeling failover spend prevents reliability improvements from blowing the budget.
How to use it
- Enter token volumes and list prices for your primary provider model.
- Enter equivalent volumes and prices for your secondary failover model.
- Set expected failover rate (percentage of requests hitting secondary).
- Review the blended bill vs primary-only baseline at your failover rate.
- Stress-test with outage scenarios at 5%, 10%, and 100% failover for a day.
- Confirm secondary model capability parity so failover does not trigger quality retries.
Planning tips
- Keep failover models in the same capability tier to avoid silent quality degradation.
- Sync prompt templates across providers—retry loops on format errors multiply failover cost.
- Primary cache savings do not transfer to secondary—expect full input price on failover path.
- Read the primary vs failover dual scales before setting an SRE outage budget.
- Run game-day drills to measure actual failover traffic, not guess outage duration.
- Negotiate credits for primary outages separately—this tool models technical cost only.
Frequently asked questions
What failover rate should I assume for planning?
For steady-state budgeting, 1–3% is reasonable if failover is rare. For outage scenario planning, model 100% secondary for the expected outage window.
Are secondary provider prices official?
CentsPerToken uses approximate list prices for planning. Verify both providers' current rates officially.
Does failover affect output token pricing asymmetry?
Yes. If secondary output rates are higher, failover during output-heavy traffic is especially expensive.
Should I include data transfer or egress costs?
This calculator covers LLM API token pricing only. Add infrastructure costs for multi-provider architecture separately.