LLM switching payback by monthly spend & new-vendor discount
Short answer: whether it pays to migrate off your current model vendor comes down to how much you spend now versus the one-time cost of rewriting prompts, evals, and eating a little downtime. Across this 42-cell sweep with a fixed migration effort, 17 of 42 spend/discount combinations pay back within 6 months — a clear switch — while 11 take over two years, meaning the lock-in cost dwarfs the savings and you should stay put. At $6,000/mo spend with a 25% discount the switch repays in 6.05 months; at $1,500/mo with only a 15% discount it takes 39.69 months. The full grid is below.
Switching payback in months by spend and discount
Rows: current monthly model spend. Columns: discount the cheaper vendor offers. Each cell is the payback period in months. Green ≤ 6, grey 6–12, amber 12–24, red over 24.
| Spend \ discount | 10% | 15% | 20% | 25% | 30% | 40% | 50% |
|---|---|---|---|---|---|---|---|
| $500/mo | 177.93 | 118.62 | 88.97 | 71.17 | 59.31 | 44.48 | 35.59 |
| $1,500/mo | 59.53 | 39.69 | 29.77 | 23.81 | 19.84 | 14.88 | 11.91 |
| $3,000/mo | 29.93 | 19.96 | 14.97 | 11.97 | 9.98 | 7.48 | 5.99 |
| $6,000/mo | 15.13 | 10.09 | 7.57 | 6.05 | 5.04 | 3.78 | 3.03 |
| $12,000/mo | 7.73 | 5.16 | 3.87 | 3.09 | 2.58 | 1.93 | 1.55 |
| $25,000/mo | 3.89 | 2.59 | 1.94 | 1.55 | 1.3 | 0.97 | 0.78 |
Provenance
- Engine
- LLM Vendor Lock-In Cost (
llm-vendor-lock-in-cost) - Source
- Computed live from
/engines/llm-vendor-lock-in-cost.js - Grid
- 6 spend levels × 7 discounts = 42 cells
- Computed
- 2026-05-23
- Held constant
- prompt complexity 5/10 · 60-test eval suite · 24 retraining hours · 1 downtime day · $120/hr engineering
Every value above and in the table is the deterministic return value of the shipped engine bundle, recomputed independently in continuous integration and diffed against this page on every build. The engine is pure (no clock, no randomness): the same inputs always produce the same output. No number on this page was hand-typed or estimated.
How to read this dataset
- Find your current monthly model spend on the left, the discount your alternative vendor offers across the top, and read the switching payback where they meet.
- Red cells are real lock-in: the migration cost takes over two years to recoup, so a marginal price cut is not worth re-validating your stack. Wait for a bigger discount or a forced reason to move.
- The CSV adds total engineering hours, the dollar switching cost, downtime opportunity cost, and monthly savings for every cell, so you can plug in your own migration estimate.
Want to run your own inputs instead of this grid? Use the LLM Vendor Lock-In Cost calculator. For the exact math, defaults, and what the model can't tell you, read the methodology page. For a worked move off a closed model, see switching from Claude to open source, and for a wider risk lens, evaluating LLM vendor risk for solo SaaS.