Quick answer
OpenRouter documents rate-limit scope as Account, model, and upstream-provider dependent. Routing cannot guarantee identical capacity across different upstream model providers.
Verified Aug 22, 2026Official source
Current limits
| Constraint | Current value | Scope | Verified source |
|---|---|---|---|
| rate-limit scopeRouting cannot guarantee identical capacity across different upstream model providers. | Account, model, and upstream-provider dependent | All accounts | OpenRouterAug 22, 2026 |
Why does this limit matter?
A fallback route can have a different rate, latency, context, or availability profile from the preferred route.
This value is scoped to All accounts; a different plan, runtime, model, endpoint, region, or account can produce a different effective constraint.
What should you check?
- Inspect the response provider, model metadata, and account usage rather than hardcoding one global RPM.
- Confirm the exact plan, model, runtime, endpoint, region, and account that serve the failing workload.
- Record the observed value, response headers or configuration, timestamp, and source without logging secrets.
Important caveats
- Free model allowances are distinct from paid-provider throughput and per-request limits.
- Treat the official source and live account configuration as authoritative if they differ from this verified snapshot.
HyperObserve reports the documented platform constraint. Your application, SDK, gateway, provider, region, or account can impose a lower effective limit.
Related