BBramaDocs
Browse documentation

Featherless featherless

Featherless's hosted open models behind an OpenAI-compatible endpoint. Routes are featherless/<model> over openai-chat, paid by the deployment's own API key.

How we authenticate

There is no login flow here. Featherless issues a long-lived API key in its own console; an operator stores it once in the Skarbiec vault, and at final use the gateway redeems the deployment's capability for this provider through the local entitlements router, reduces the stored document to the bare key, and sends it as Authorization: Bearer <key>. The key has no refresh path and no invented expiry — it works until Featherless revokes it or an operator replaces it in the vault.

The two failure shapes point at two different repairs: a 503 credential_unauthorized means no capability, read grant or trust material produced a key at final use — repair the authorization chain (runbook); a 502 provider_failure with a provider_authentication: prefix means the key was produced and Featherless rejected it — repair the provider account.

Models

One pinned catalog row, widened (never replaced) by live discovery through the models endpoint: featherless/TheDrummer/Cydonia-24B-v4.3.

Plan usage

Featherless publishes no free plan-usage report for API keys. The ledger records that absence as the provider's own answer, never as a zero (subscription).

Endpoints

Chathttps://api.featherless.ai/v1/chat/completions
Model discoveryhttps://api.featherless.ai/v1/models
Auth headerAuthorization: Bearer <key>
Trusted hostapi.featherless.ai
Base URL overrideBRAMA_PROVIDER_FEATHERLESS_BASE_URL, validated with the refusals in provider