Troubleshooting

`401 invalid_api_key`

Key was revoked (Workspace → Virtual keys → Audit shows the event).
Key is from a different environment (sk_test_ vs sk_live_).
Wrong header (must be Authorization: Bearer …).

`400 model_not_allowed`

The workspace allowlist is on and the requested model isn't in it. Either add the model under Routing → Allowed models or call a model that's already allowed.

`400 unknown_model`

The slug isn't in the catalog. Check spelling. List allowed models via the dashboard.

`402 no_vendor_key`

No vault key configured for the vendor that owns this model. Workspace → Routing → Vault keys → add one.

`402 subscription_inactive`

Workspace's odnoga subscription is canceled or unpaid. Workspace → Pay odnoga.

`402 budget_exhausted`

Active window hit 100%. Wait for x-ratelimit-reset or raise the cap.

`429 rate_limited`

Vendor returned 429 and all fallback keys are cool-down. Back off using x-ratelimit-reset.

`502 upstream_error`

Vendor returned an error. Open the row in Requests — error_message carries the vendor's reason. Most common:

Anthropic credit balance is too low → top up vault key.
OpenAI model_not_found → the slug is stale; refresh catalog or pick a current one.

`504 timeout`

Vendor didn't respond in time. Retry with backoff. If a specific model is consistently slow, switch or add a fallback policy.

Headers look fine, analytics are empty

You're not setting x-airouter-end-user. Add it and traffic re-populates Top end users from the next call onward.

Streaming hangs after the first chunk

A proxy in front of your code is buffering. For Nginx: proxy_buffering off. For Vercel: stream with the correct response type.

Costs say $0 on every row

Every row is status = error — there were no tokens to bill. Look at error_code; once the call succeeds, cost populates.

Still stuck?

Grab the x-airouter-request-id from the failing call and share it with support. That row contains everything we need.