Envoy AI Gateway token usage histogram (OTel gen_ai semconv)

WithDescription('Number of tokens processed.') and WithUnit('token'). Declared as a HISTOGRAM, not a counter - which is the OTel GenAI semconv choice so that input and output token distributions can be read separately, discriminated by the gen_ai.token.type attribute (values 'input' and 'output'). If you are porting a dashboard from an engine where token counts are monotonic counters, the rate() over this is not equivalent. 14 explicit bounds from 1 token to 67 million. Envoy AI Gateway main (envoyproxy/ai-gateway), read 2026-10-05. Metrics go through an OpenTelemetry MeterProvider (internal/metrics/metrics.go NewMeterFromEnv), not a bare prometheus_client registry: a Prometheus reader is ALWAYS attached, and a console or OTLP exporter is added additionally when OTEL_METRICS_EXPORTER is 'console', or when it is neither 'none' nor 'prometheus' and an OTLP endpoint env var is set. OTEL_SDK_DISABLED=true turns off the extra exporters but not the Prometheus reader. That means the metric names are dotted OTel names (gen_ai.server.*), which the Prometheus exporter renders with dots replaced by underscores and a unit suffix appended - so the scraped series is NOT the literal constant in genai.go. Names follow the OpenTelemetry GenAI semantic conventions, not any engine's private scheme.