Observability added to AI Gateway alpha
By Steven Van ·
New Observability metrics show per-model request counts, latency, token usage, and cost across the AI Gateway alpha.
Vercel's Vercel AI Gateway, currently in alpha for all users, now has a dedicated AI section in Vercel Observability. The gateway lets developers switch between roughly 100 AI models without managing separate API keys, rate limits, or provider accounts.
The new section surfaces:
- Requests by model
- Time to first token (TTFT)
- Request duration
- Input/output token count
- Cost per request (free while the gateway is in alpha)
Metrics can be viewed across all projects or drilled into by project and model, to compare latency and estimate what requests would cost in production. Read the announcement.
Vercel
The platform for frontend developers — deploy, preview, and scale web apps and AI agents with zero config.
View Vercel →