Track every AI request. Calculate costs. Enforce budgets. Inspect traffic end-to-end.
Intercepts AI requests from Laravel's HTTP client, the OpenAI PHP SDK, and raw Guzzle clients — zero code changes required. Provider macros offer a convenient alternative with built-in authentication and base URL configuration.
Track text completions, embeddings, image generation, video generation, text-to-speech, and speech-to-text operations across all major AI providers in a single, unified observability layer.
Token, image, video, audio duration, and character-based pricing with full tier support including OpenAI batch, flex, and priority tiers as well as Anthropic batch pricing.
Define daily, weekly, and monthly spending limits per user, team, or any Eloquent model. Enforce hard blocks via middleware or fire soft warning events at configurable thresholds.
A built-in single-page application for exploring requests, drilling into payloads, comparing model costs, managing the pricing catalog, and viewing per-user analytics.
Track streaming SSE responses with time-to-first-token metrics. Optionally download and persist generated images and videos to any Laravel filesystem disk before provider URLs expire.