What it does
Odigos is an observability control plane that installs into a Kubernetes cluster and automatically instruments supported workloads. It uses OpenTelemetry and eBPF to produce telemetry without requiring application code changes, then manages collectors and exports data to compatible destinations. The open-source edition focuses on core tracing, while the vendor's Enterprise tier adds capabilities such as deeper eBPF capture, database tracing, and multi-cluster administration. Odigos is most relevant to platform teams that already operate Kubernetes and want to standardize how services emit telemetry.
Build a small pipeline before expanding it
Odigos is installed into an existing Kubernetes cluster. The public guide offers a CLI, Helm chart, or OpenShift Operator, and warns that each install method acts on the active Kubernetes context. Check that context before applying anything. The quickstart shows a local kind or minikube cluster, a multi-service demonstration app, and Jaeger as a separate trace backend. Its final viewing step port-forwards Jaeger's interface to a local browser. These are documented examples, not commands run for this listing.
That arrangement clarifies Odigos's role: it instruments workloads and routes OpenTelemetry data, while a compatible backend stores and displays the resulting traces. A sensible first exercise is to select one service, connect one destination, and confirm that the destination receives trace data before widening the rollout.
Pick the edition for the signals you need
The pricing page lists a free, Apache-licensed Open Source edition and a custom-priced Enterprise tier with a 14-day trial. The open-source tier covers core OpenTelemetry tracing; the page lists extended low-overhead eBPF capture, database tracing, and multi-cluster administration among Enterprise additions. Those boundaries matter if a team expects the free edition to supply the more advanced collection and fleet-management features. The exact Enterprise quote is not published on the pricing page.
Because Odigos exports OpenTelemetry data, its output can be sent to compatible managed or self-hosted destinations rather than locking the team into an Odigos storage backend. The tradeoff is operational: the team still needs a Kubernetes environment and a destination for the telemetry.
Strengths
- Automatically instruments supported workloads without requiring application code changes.
- Manages collector deployment and scaling, and exports telemetry in OpenTelemetry format to compatible backends.
- Offers a free self-hosted Open Source edition alongside a separate Enterprise tier.
Watch out for
- Installation requires an existing Kubernetes cluster, and the documented methods target the active Kubernetes context.
- The pricing page reserves extended eBPF capture, database tracing, and multi-cluster administration for Enterprise.
Best for
- Platform teams adding distributed tracing to supported services without changing application code.
- Kubernetes operators who want Odigos to manage collection and export OpenTelemetry data to an existing backend.
- Teams building a local tracing demonstration with a Kubernetes cluster and a separate Jaeger backend.
Sources
- Odigos | The Production Context Platform (Oct 26, 2026)
- Overview - Odigos (Oct 26, 2026)
- Installation - Odigos (Oct 26, 2026)
- Introduction - Odigos (Oct 26, 2026)
- Viewing traces - Odigos (Oct 26, 2026)
- Pricing | Odigos (Oct 26, 2026)
- odigos-io/odigos (Oct 26, 2026)
- LICENSE (Odigos) (Oct 26, 2026)