Summary
Vercel AI Gateway now produces an OpenTelemetry trace for every request, and Pro and Enterprise teams can stream those traces to any OTLP/HTTP endpoint via Vercel Drains. Traces capture routing, retries, token usage, and latency while excluding prompt and completion content.
What changed
On August 5, 2026 Vercel added OpenTelemetry tracing to AI Gateway. Every request generates a trace covering routing decisions, retries, token consumption, and performance; teams export through Drains to OTLP/HTTP endpoints at $0.05 per 1,000 traces, controlling volume with sampling.
Why it matters
Observability is a persistent gap in LLM infrastructure. Standardized OpenTelemetry traces let teams debug latency, retries, and routing across model providers using existing tooling such as Datadog, Honeycomb, and Grafana instead of provider-specific dashboards, lowering the cost of running a gateway in production.
Evidence excerpt
AI Gateway now produces an OpenTelemetry trace for every request. Pro and Enterprise teams can send these traces through Vercel Drains to any OTLP/HTTP-compatible endpoint.