Errors and what to do

This page lists the messages the LLM Gateway returns when it refuses a call, and how to fix each one. When the gateway does not send a call to a provider, it answers in the caller’s own format (Anthropic, OpenAI or Gemini), so your coding tool or app shows the message as an error. The call costs nothing, and it appears in Logs with the reason in Status and in its Gateway decisions. In Xpander Chat, a run refused this way ends with a card titled xpander did not send this request, followed by the same message and what to do.

Your key

Status	Message	What to do
`401`	This credential is not an active xpander key.	The key is wrong, revoked, or missing from the call. Check the key your coding tool sends, or create a new one in AI models & keys.
`401`	This xpander key has expired.	Create a new key in AI models & keys.
`400`	Name the model to call in the request’s model field.	Set `model` in the call.

The model

Status	Message	What to do
`403`	This model is not enabled for you on the xpander LLM Gateway.	Pick a model listed in AI models & keys, or ask an admin to enable it in Models. A model served on your organization’s own endpoint is called as `<provider ID>::<model>`, the name `/v1/models` lists.
`403`	Model calls for this model or this account are paused by your xpander admin.	An admin paused the model. Pick another model, or ask your admin.
`403`	Your organization’s LLM Gateway settings do not allow this kind of provider access for you.	The license model for your group or agent does not allow how this call would be paid. Ask your admin.

Provider keys and endpoints

Status	Message	What to do
`403`	No provider key is set up for this model in your xpander organization.	An admin adds a key for the model’s provider in Models (Providers and keys). On a Hybrid or Air-Gapped install, every provider your people use needs a key of your own.
`403`	Your organization has no … key on the xpander LLM Gateway for this model.	An admin adds a key for the model’s provider in Models (Providers and keys).
`403`	Your organization has no … key stored in xpander for this model. Tools on your computer use the keys stored in xpander, not your own environment’s credentials.	For people of a Hybrid organization: their coding tools connect through xpander cloud, which cannot use the credentials inside your install. An admin adds a key for the provider in Models.
`403`	The model provider refused the key your organization set up for this model.	The provider rejected the stored key (revoked, expired, or without access to the model). An admin replaces it in Models.
`403`	Your organization’s … key is stored without its endpoint.	An Azure key needs the endpoint it belongs to. An admin adds it next to the key (Azure endpoint). Any other model works meanwhile.
`403`	The endpoint configured for this model’s provider key is not a reachable provider address.	The endpoint points at an address the gateway does not call, such as a private address on xpander cloud. An admin sets the endpoint’s public `https` address (Model endpoints).

On xpander cloud, saving an endpoint at a private, loopback or cloud metadata address fails at once with Model endpoints are called at public addresses here, so this endpoint was not saved.

Budgets, credits and rate limits

Status	Message	What to do
`429`	The daily / weekly / monthly spending limit set for this … on the xpander LLM Gateway is reached until …	A budget set to Block is spent. Wait for the time shown, or ask an admin to raise it.
`402`	Your xpander organization reached its spending limit, so model calls are paused.	The organization limit is reached. An admin raises it in Budgets.
`402`	Your xpander organization is out of credits, so model calls are paused.	An admin adds credits in billing.
`403`	Your xpander account is paused, so model calls are not available.	Contact your xpander admin.
`403`	Your xpander organization is not active, so model calls are paused.	An admin reactivates the organization.
`429`	The request rate set for this … is reached.	Retry in a minute. Coding tools and SDKs retry this on their own.
`429`	Too many calls are running at once for this …	Retry when one finishes. Coding tools and SDKs retry this on their own.

Your own Claude plan

Status	Message	What to do
`403`	Your own Claude plan is not enabled for this call on the xpander LLM Gateway …	The message names what is missing: your organization does not allow your own plan for you, the call is not from Claude Code, or the model is not a Claude model from Anthropic. Sign out of the plan in Claude Code to use company models, or ask your admin.
`429`	Your Claude plan’s usage limit is reached until …, and plan calls are set to wait for the plan rather than use company models.	Your plan is at its limit and your calls pause until it resets. Wait, or choose Continue on company models in AI models & keys when your organization allows it.
`403`	Your Claude login expired.	Run `/login` in Claude Code, then retry.
`403`	Anthropic did not allow this call on your Claude plan.	The model or feature is not part of your plan. Pick another model.

The gateway itself

Status	Message	What to do
`503`	The LLM Gateway could not load your organization’s settings just now.	Retry in a few seconds.
`503`	The LLM Gateway could not verify this key just now.	Retry in a few seconds.
`503`	The LLM Gateway could not load the provider key for this model just now.	Retry in a few seconds.
`404`	The LLM Gateway serves model calls only (messages, chat completions, responses, embeddings, token counting).	Call a model path: `/v1/messages` (and `/v1/messages/count_tokens`), `/v1/chat/completions`, `/v1/responses` or `/v1/embeddings`. `/v1/models` lists the models you may call.
`403`	This call was routed to a model outside your organization’s settings, so it was not sent.	Retry the call.

When the gateway is unavailable

Agents call models only through the LLM Gateway. When it cannot be reached for 30 seconds or more:

Xpander Chat shows a banner: AI gateway is unavailable. Agents can’t run right now. We’ll reconnect automatically.
A new run or chat message is refused at once, and a run in progress stops at its next model call, with the card This run stopped because the AI gateway is unavailable. Retry once it is back.
Calls from coding tools and SDKs answer 503 with a Retry-After header.

In the meantime, nothing is sent to a provider any other way. The gateway reconnects on its own, usually within a minute, and the banner clears.

When the provider refuses

When the call reached the provider and the provider refused it (for example, an invalid request or a model that does not exist), the provider’s own error comes back unchanged. In Xpander Chat the card is titled The model provider rejected this request.

Govern

LLM Gateway

Deploy

Integrations

Licensing and identity

Security and compliance

Errors and what to do

Your key

The model

Provider keys and endpoints

Budgets, credits and rate limits

Your own Claude plan

The gateway itself

When the gateway is unavailable

When the provider refuses

​Your key

​The model

​Provider keys and endpoints

​Budgets, credits and rate limits

​Your own Claude plan

​The gateway itself

​When the gateway is unavailable

​When the provider refuses

Your key

The model

Provider keys and endpoints

Budgets, credits and rate limits

Your own Claude plan

The gateway itself

When the gateway is unavailable

When the provider refuses