Skip to main content

On every response

On /v1/messages the same request-window figures are also sent under Anthropic’s names (anthropic-ratelimit-requests-limit, -remaining and -reset), so Anthropic SDKs read them unchanged. Where a plan has a tokens-per-minute limit, x-ratelimit-*-tokens (and anthropic-ratelimit-tokens-*) report it the same way.
Example response headers
The x-ratelimit-* trio lets a client pace itself instead of discovering the limit by hitting it: throttle as remaining approaches zero. Per-plan limits are published on Plans & limits.

On error responses

Honor retry-after exactly; the value is computed, not a guess. The retry recipe (and which statuses to retry at all) is on Errors.

Request headers

Besides the authentication header (Authentication), these request headers are part of the contract:

Legacy spellings

sideren-model is also sent as x1-model with the same value, and x-sideren-payment-mode also accepts the older x-layerx1-payment-mode. Read either one; new code should use the sideren names.
This is the customer-visible header surface. Anything else you may observe is not part of the API contract and can change without notice: build only against the headers on this page.