Providers · Beta

Govern Fireworks AI traffic before it reaches a shared model endpoint.

First-class Fireworks BYOK with the official inference base, serverless catalog verification, and Chat Completions eligibility.

Start in a test environment Read current API behavior

How the connection works

Provider access stays separate from application control.

Check serverless eligibility explicitly

Fireworks distinguishes serverless and dedicated deployments, and not every model is serverless. The gateway connection check reads the account catalog and requires the full model identifier instead of treating every visible model as an interchangeable public endpoint.

Bound shared-endpoint traffic

A workload can carry its own dollar budget, request and token rates, concurrency ceiling, and compatible fallback. Those limits contain a burst before it consumes a shared provider allowance or multiplies retries across targets.

Normalize usage for reconciliation

Chat-completion responses carry token usage that the gateway records at micro-dollar precision with the selected route and provider target. Small requests are accumulated rather than rounded to a zero-dollar operating history.

Current shared controls

Status determines what can be configured.

Write-only, encrypted BYOK for implemented provider connections Stable model aliases and protocol-eligible route targets Organization and key budgets, RPM, TPM, and concurrency hard limits Observe, Shadow, Enforce, reason codes, and metadata-only evidence

Beta connections require customer validation against the exact model, payload, streaming mode, region, and provider account before production use.

Client boundary

AIGHQ_BASE_URL=https://api.aigatewayhq.com/v1
AIGHQ_API_KEY=from-your-workspace
AIGHQ_MODEL=company-approved-fast

Applications keep a stable gateway URL and model alias while an administrator changes eligible provider accounts and models.

Implemented boundary

Only documented protocols become eligible.

Choose Fireworks AI, enter the write-only API key, verify access to the serverless catalog, and use the full Fireworks model identifier in the route.

POST /v1/chat/completions GET /v1/accounts/fireworks/models connection test

Compatibility is bounded to the provider's current published interface. Review the provider documentation before approving a production model.