Skip to content

phoenix

varize-phoenix-v17.11.0 scope: arize-phoenix Feature

This release adds 4 notable features for engineering teams evaluating rollout.

Published 1mo Tracing
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agents ai-monitoring ai-observability aiengineering anthropic datasets
+10 more
evals langchain llamaindex llm-eval llm-evaluation llmops llms openai prompt-engineering smolagents

Summary

AI summary

Bug fixes, features across agent, annotations, evals, server modules in version 17.11.0 released on 2026-06-24.

Full changelog

17.11.0 (2026-06-24)

Features

  • agent: submit highlighted command on Enter in skill prompt (#13868) (ea02242)
  • annotations: show trace annotation summaries in project stats panel (#13855) (604a087)
  • evals: agent-readable failure reports for PXI eval harness (closes #13668) (#13737) (c59fb82)
  • server: expose agent endpoints in Swagger UI when running with --dev (#13873) (25cb3de)

Bug Fixes

  • server: cap pydantic-ai-slim below 2.0 to unbreak PyPI installs (#13881) (379e231)
  • trace: add bounded HTTP timeouts to dataset/trace fixtures (#13747) (6438a80)
  • ui: restore outline shape for thumbs up/down feedback icons (#13871) (e3bc988)

Documentation

  • tutorials: level up email model-comparison into eval-harness cookbook (DEV-357) (#13718) (537b7e6)
  • tutorials: level up session-level evaluation cookbook (#13728) (d411ea0)
  • tutorials: level up the realtime guardrails cookbook (concept-first) (#13721) (ef2be08)

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

Share this release

Track phoenix

Get notified when new releases ship.

Sign up free

About phoenix

AI Observability & Evaluation

All releases →

Related context

Earlier breaking changes

Beta — feedback welcome: [email protected]