OpenRouter 402 in_flight_budget_exhausted: what it means and how to fix it

What it means

Your running and recently completed paid requests already fill your in-flight spending budget, so the request is rejected before reaching a provider even though the balance is positive. metadata.limit_source is openrouter_in_flight_budget. This is transient and the response includes a Retry-After header.

Causes

• Many concurrent paid requests whose estimated cost fills the in-flight budget
• Recent request costs still settling

How to fix it

1. 1
   Fix 1.

   Wait for the Retry-After header and retry explicitly; SDKs do not retry 402 on their own.

   Snippet

   Copy

   const res = await fetch('https://openrouter.ai/api/v1/chat/completions', { ... });
   if (res.status === 429 || res.status === 503 || res.status === 402) {
     const retryAfter = Number(res.headers.get('Retry-After'));
     if (Number.isFinite(retryAfter) && retryAfter > 0) {
       await new Promise((r) => setTimeout(r, retryAfter * 1000));
       // retry the request
     }
   }

2. 2
   Fix 2.

   Adding credits raises the in-flight budget, up to a capped ceiling.