Model Fallback

Simple Definition

A model fallback is a backup AI model or provider that can take over when your main model is unavailable, too expensive, rate-limited, or simply not good enough for a task.

It is your backup plan for AI. If your first model fails, another model can still complete the work.

Why It Matters

Having a tested fallback turns an outage or price spike from a crisis into a minor inconvenience. The key word is tested — a backup only helps if you have confirmed it can actually do the job before you need it.

Example

If Claude is unavailable, a team switches the same task to ChatGPT, Gemini, Codex, or a local model — and keeps working with little disruption.

Related Terms

• AI Model Dependency Risk — the risk a fallback is designed to reduce
• AI Model Access Risk — the business risk a fallback protects against
• Model Routing — sending work to different models, which can include fallbacks
• AI Workflow — the process your fallback keeps running
• Local LLM — a self-hosted model that can serve as a fallback