Skip to content
Agent Radar

Phoenix 20.23 adds OTLP server-log export

Updated Phoenix

Summary

Phoenix’s October 16 update exports server logs over OTLP and fixes Bedrock tool-result handling in the playground.

Tool
Phoenix
Category
Agent infra
Date
Open source
No
License
Elastic-2.0
Pricing
Repository licensing does not include hosted-service, model-provider or infrastructure costs. No service prices are claimed in this release note.
Sources

Arize published Phoenix 20.23.0 on October 16, 2026.

What changed

The new server feature exports logs over OTLP. This gives operators a documented release-level reason to inspect how Phoenix server logs fit their telemetry pipeline; it is not a claim that export is enabled automatically in every deployment.

The playground also fixes how Bedrock tool results are forwarded, sending them directly as text. Documentation in the release adds an Atomic Agents tracing integration.

Scope

This is an observability-tool update, not a model launch. The release does not publish independent performance measurements or establish that every hosted plan includes the feature. For deployment decisions, distinguish the software's license from a hosted service's commercial terms.

License and cost

The release-tag license is Elastic License 2.0, including a restriction on offering substantial functionality as a hosted or managed service. We therefore describe this as source-available rather than unrestricted open-source software. Software licensing is separate from hosted-service, inference and infrastructure charges; this note does not claim a free hosted tier.

These changes are reported by the maintainers. We have not independently benchmarked the release.

Explore the Phoenix hub and agent infra category.

More from Phoenix

View all

  • Phoenix v20.21.0 adds Mistral AI cost tracking

    PhoenixArize Phoenix v20.21.0, released October 11, 2026, adds Mistral AI model pricing to cost tracking and fixes evaluator metadata loss.

Questions

What is Phoenix?

The GitHub repository Arize-ai/phoenix is described as "AI Observability & Evaluation". Its docs list features such as LLM-based evaluations, prompt management, a prompt playground, span replay, datasets and experiments.

Is Phoenix open source?

Not stated in the cited sources. The excerpts say the GitHub API does not report a recognized license for the repository, so no license is claimed here.

How is Phoenix used?

According to its docs, you can create datasets by collecting traces or uploading from code/CSV, run experiments to compare different versions of your application, and attach evaluators to datasets so they run automatically during experiments.

Filed on Agent Radar. Claims in this note are limited to the sources above.