Arize Phoenix vs Weights & Biases Weave

Pick two tools from the same category to compare side-by-side.

	Arize Phoenix Open-source LLM tracing and evaluation platform from Arize AI, built on OpenTelemetry.	Weights & Biases Weave LLM tracing and evaluation toolkit from Weights & Biases for developing generative AI applications.
Category	observability	observability
Vendor type	hybrid	hybrid
Pricing	freemium	freemium
Strengths	+ OpenTelemetry-based instrumentation works across frameworks and providers + Built-in evals, datasets, and experiments alongside tracing + Self-hostable via pip, Docker, or Helm in addition to hosted cloud + Prompt playground for replaying and comparing traced LLM calls	+ Minimal instrumentation via the weave.op decorator + Evaluation framework for structured, comparative LLM assessments + Integrates with the broader Weights & Biases ML experiment platform + Apache-2.0 licensed client SDK
Tradeoffs	− Elastic License 2.0 (ELv2) restricts offering Phoenix itself as a managed service − Self-hosting means operating and scaling the collector and storage yourself − Advanced enterprise monitoring features live in the separate commercial Arize platform	− Tied to a W&B account and backend rather than a standalone self-hosted store − TypeScript SDK is less mature than the Python SDK
Used in stacks	0	0