Describe what Arize Phoenix is and where it diverges from Arize AX

Arize is the company. They make two products that share the same core model and the same OpenInference attribute conventions. **Phoenix** is the open-source, self-hostable tracing and evaluation tool. You can run it locally with `pip install arize-phoenix` and `phoenix.launch_app()` or as a container in your infrastructure. It is OTel-native (it speaks OTLP) and has strong features for LLM-specific work: visualizing the call tree of an agent, retrieval-quality views for RAG (which chunks were retrieved, with what scores), and LLM-as-judge evaluators. It is free. **Arize AX** is the hosted enterprise platform. It has the same conceptual model, the same OpenInference shape and the same kind of tracing UI. AX layers on top: production-scale storage, longer retention windows, team features (RBAC, shared dashboards), ML monitoring for traditional non-LLM models (tabular model drift, fairness metrics), SLAs, and enterprise support. The relationship: Phoenix is the starting point for most teams. AX is where they move when volume, retention, compliance, or team-collaboration requirements outgrow self-hosting.

Arize Phoenix and Arize AX are an open source plus hosted enterprise pair from one company. They share data model and instrumentation conventions but diverge on scale, deployment, and breadth. **Phoenix is the OSS tool.** It is ELv2-licensed (Elastic License 2.0) and installs as a Python package or Docker container. It speaks OTLP and the OpenInference semantic conventions (Arize's LLM-specific attribute spec, broadly compatible with OTel gen_ai semconv). Core features are LLM-tracing-first: an agent and RAG-aware UI that visualizes the call tree, retrieval-quality views (per-query top-k with scores and content previews), and a built-in LLM-as-judge evaluator framework. It runs on a developer laptop, in a self-hosted container, or as a sidecar in Kubernetes. **Arize AX is the hosted enterprise platform.** It shares Phoenix's data model (traces, evals, datasets, scores) and instrumentation: AX displays OpenInference-shaped OTLP spans with the same views. The differences are operational: - Production-scale storage and longer retention (months to years). - ML monitoring for non-LLM models (classical tabular models, drift detection, fairness metrics, feature attribution). - Team features: RBAC, audit logs, shared dashboards across projects, alerts to PagerDuty and Slack. - SLAs and enterprise support. - Compliance certifications (SOC 2, HIPAA-eligible deployments). **The migration path.** Teams start with self-hosted Phoenix to validate instrumentation and use cases. When trace volume exceeds what self-hosting comfortably handles, retention requirements grow, or the team needs broader observability (ML models alongside LLM agents), they move to AX. Migration is low-friction: both consume the same OpenInference-shaped spans, so instrumentation code does not change.