Skip to documentation

Build a free, local, and paid model order

Build a free, local, and paid model order

Create an explicit model order that uses free or local capacity before a paid fallback.

Purpose

Create a Standard Group that prefers selected free or local capacity and uses a paid model only when the wait budget permits fallback.

Relay follows the order that you configure. Relay does not discover the cheapest model or optimize this order automatically.

Prerequisites

Add and verify these model types:

  1. A free or subscription-backed model with known capacity.
  2. A local OpenAI-compatible model.
  3. A paid API model.

Add configured pricing if you want Relay to estimate cost or enforce spend limits.

Create the order

  1. Open Groups.
  2. Add a Standard Group named free-local-paid.
  3. Enable fallback.
  4. Set the maximum wait that the caller accepts.
  5. Add the free model at rank 1.
  6. Add the local model at rank 2.
  7. Add the paid model at rank 3.
  8. Save the group.

Add capacity controls

  1. Add the known request or token limit to the free model.
  2. Enable request pacing for second- or minute-level request limits when smooth spacing is useful.
  3. Add a concurrency limit to the local model if local hardware has a fixed parallel capacity.
  4. Add a maximum-latency value when a hanging upstream must be cancelled.

Understand selection

ConditionRelay action
Rank 1 is eligible nowSelect rank 1
Rank 1 becomes eligible inside the wait budgetWait for rank 1
Rank 1 exceeds the budget and rank 2 fitsSelect rank 2
Ranks 1 and 2 exceed the budget and rank 3 fitsSelect rank 3
No candidate fitsReturn 429 with Retry-After

An upstream 429 for an active task cools the selected model and requeues that same task. It does not automatically move that task to the next rank.

Expected result

Relay prefers the configured free or local Models. It selects the paid Model only when the higher-ranked Models cannot serve inside the wait budget.

Verification

Use Realtime to confirm each model's limit, cooldown, and selected state. Use Logs to confirm the group, candidate evaluation, selected route, wait time, and cost.

Recovery

If the paid model receives traffic too early, increase the group wait budget or inspect the higher-ranked models' limits and cooldowns.

Next

Tune wait budgets, pacing, and cooldowns.