Monitoring & Observability for every AI request.
Real-time logs, metrics, and tracing out of the box. Broadcast traces to your preferred observability platform with zero code changes.
How observability works
EUrouter captures structured traces for every request and lets you forward them to your existing tools. No SDKs to install, no code to change.
Real-time request logs
Every request generates a structured trace with status codes, latency, token usage, cost, model, and provider. Filter and explore your traffic directly from the dashboard.
Read more
Broadcast to your tools
Forward traces to Langfuse, Datadog, PostHog, LangWatch, any OpenTelemetry Collector, or a raw webhook endpoint. No SDK changes needed.
Read more
Privacy and configuration
Control what gets sent and where. Strip sensitive content, adjust sampling rates, and route traces from specific API keys to specific destinations.
Read more
Real-time request logs
Every API request generates a structured trace in real time. See status codes, latency, token usage, cost, and the provider that served each request from your dashboard.
- Filter logs by model, API key, status code, or time range
- View token counts broken down by input, output, and total
- See which provider, routing strategy, and fallback count were used
- Track cost per request in EUR with provider-level pricing
Broadcast to your tools
Broadcast forwards traces from your EUrouter requests to external observability platforms. Connect to Langfuse, Datadog, PostHog, LangWatch, any OpenTelemetry Collector, or a raw webhook endpoint.
- Traces are sent asynchronously, adding zero latency to your responses
- No SDK changes or additional instrumentation in your code
- Connect multiple destinations simultaneously
- Test connections before saving to verify credentials
Supported destinations
Connect to your preferred observability platform in minutes. No SDK changes, no additional instrumentation.
Langfuse
langfuse.comOpen-source LLM engineering platform for tracing, evaluation, prompt management, and datasets. Self-host or use their cloud offering.
LangWatch
langwatch.aiLLM observability and analytics platform with guardrails, quality scoring, and cost tracking built in.
Datadog
datadoghq.comEnterprise cloud monitoring with APM, dashboards, alerting, and LLM observability. Ideal for teams already using Datadog for infrastructure.
PostHog
posthog.comProduct analytics platform with LLM observability, session replay, feature flags, and A/B testing. EU cloud available.
OpenTelemetry Collector
opentelemetry.ioSend traces to any OTLP-compatible backend including Grafana Tempo, Jaeger, Honeycomb, Axiom, and SigNoz.
Webhook
Raw OTLP JSON delivered to any HTTPS endpoint you control. Build custom integrations or pipe traces into your own systems.
10%Privacy and configuration
Configure privacy, sampling, and key filtering independently per destination. Strip sensitive content from traces, reduce volume on high-traffic workloads, or isolate monitoring by environment.
- Privacy mode strips prompt and completion text before forwarding
- Sampling rate controls what percentage of traces are sent
- API key filter routes traces from specific keys to specific destinations
- All destination credentials are encrypted before storage
Integrate AI without GDPR risk.
You need AI that won’t create compliance headaches. Your data stays in the EU, GDPR is enforced by default, and every request is routed for the best balance of cost, latency, and uptime, reducing risk while improving performance.