Monitor Fireworks AI with Respan
Overview
Respan routes Fireworks AI model calls through the gateway, providing unified observability including logs, cost tracking, latency metrics, and reliability monitoring.
Configure your Fireworks API key globally in the Respan dashboard with optional load balancing weights, or pass credentials per-request for multi-tenant applications.
How it works
Add your Fireworks API key in the Respan Providers dashboard. Optionally configure multiple keys with load balancing weights for redundancy.
Route requests through the gateway or log direct Fireworks SDK calls asynchronously. Both approaches provide full cost, latency, and token tracking.