LLM Vendor Lock-In Cost

How to use it

1. Enter your current monthly LLM spend, your prompt complexity on a 1-10 scale, the size of your eval suite, the engineering hours you expect retraining and reintegration to take, the days of downtime during the switch, the monthly revenue that stops while the feature is down, your hourly engineering cost, and any discount the new vendor offers. Prompt complexity (about 4 hours per point) and eval-suite size (about half an hour per test) drive the rewrite estimate, because the more your prompts are tuned to one model's quirks, the more engineering it takes to make them work on another.
2. Read the total switching cost, the engineering hours behind it, how many months of current LLM spend it equals, and the payback period at the new vendor's discount; the chart splits the total into engineering cost and downtime cost. The total switching cost is the real price of the migration, and the months-of-spend figure puts it in terms you already understand.
3. Use the payback period as the decision gate. If a 30% discount takes years to repay the switching cost, the lock-in is effectively permanent and the discount is a distraction; if it pays back in a few months, the migration is worth scheduling. This reframes vendor choice from a sticker-price comparison into a real return-on-investment question.
4. Treat the engineering and downtime estimates as the part most teams underestimate. Prompts that were tuned over months, evals that encode hard-won edge cases, and the revenue lost during a cutover are all real costs that a headline price-per-token comparison ignores entirely. Model a pessimistic case with more rewrite hours and an extra day of downtime to see how quickly the payback period stretches.
5. Re-run whenever a vendor changes pricing or a new model makes switching tempting, and use the output to negotiate. Knowing your true switching cost is leverage with your current vendor, since the discount a competitor must offer to clear your payback hurdle is exactly the number your incumbent has to beat to keep you.

The same numbers as a table

• Reference LLM switching payback by monthly spend and new-vendor discount

  Precomputed by this calculator across common inputs, to cite or download as CSV.

Related calculators

• Make the call Build vs Buy Decision Engine

  Compare building infrastructure yourself versus buying managed services with per-component build/buy verdicts.

• Make the call One-Person SaaS Valuation

  Estimate what your solo SaaS is worth using indie/micro-SaaS multiples with key valuation factors.

• Make the call Ship-or-Kill Decision Score

  Get a brutally honest SHIP, ITERATE, or KILL verdict on your side project based on traction and market signal.

• Make the call Pricing Model Picker

  Flat monthly, per-seat, usage-based, or hybrid? Compare projected revenue side-by-side.

Articles

• 9 min read LLM Vendor Lock-In Cost: Claude to Open-Source, Priced

  Price the migration from Claude to open-weights LLMs end-to-end: engineering hours, downtime, payback months. Below $1,500/mo spend, the math does not work.

• 13 min read Vendor Lock-In Math: The Real Cost of Switching LLMs

  Vendor Lock-In Math: the Real Cost of Switching LLMs: switching from Anthropic to OpenAI in production costs $15k-$200k across five buckets: API integration.

• 11 min read Model Price Shock vs Vendor Lock-In: Two Defenses Compared

  Model price shock vs vendor lock-in: a 200% spike against the cost of switching vendors. The cheaper defense is rarely the one founders pick.

More in AI product economics

Understand the costs, margins, and pricing of building AI-powered products.

• Startup Cost Estimator
• Profit Margin Calculator
• Profit Margin / Markup / Discount Calculator
• SaaS Pricing Strategy Calculator

By situation

• Value your business

Use it from code

The same calculation as a JavaScript module you can import. It runs where you import it, with no request, key or rate limit.

import { compute } from "https://aibizhub.io/engines/llm-vendor-lock-in-cost.js";

Input and output contract, TypeScript types and the guide for agents.