Skip to content

phoenix

varize-phoenix-v17.5.0 scope: arize-phoenix Feature

This release adds 3 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

Updates Bug Fixes, 17.5.0, and 2026-06-12 across a mixed release.

Changes in this release

Feature Medium

add a subagents toggle to assistant settings

add a subagents toggle to assistant settings

Source: llm_adapter@2026-06-13

Confidence: high

Feature Medium

improve agent product knowledge

improve agent product knowledge

Source: llm_adapter@2026-06-13

Confidence: high

Feature Medium

pick a time range from a calendar in the time range selector

pick a time range from a calendar in the time range selector

Source: llm_adapter@2026-06-13

Confidence: high

Dependency Low

update arize-phoenix-client to 2.8.0

update arize-phoenix-client to 2.8.0

Source: llm_adapter@2026-06-13

Confidence: high

Dependency Low

update arize-phoenix-evals to 3.1.0

update arize-phoenix-evals to 3.1.0

Source: llm_adapter@2026-06-13

Confidence: high

Bugfix Medium

add Anthropic computer-use beta header

add Anthropic computer-use beta header

Source: llm_adapter@2026-06-13

Confidence: high

Bugfix Medium

focus PXI input on open

focus PXI input on open

Source: llm_adapter@2026-06-13

Confidence: high

Bugfix Medium

don't mask GraphQL syntax and validation errors

don't mask GraphQL syntax and validation errors

Source: llm_adapter@2026-06-13

Confidence: high

Bugfix Medium

expire prompt tool diffs on provider change

expire prompt tool diffs on provider change

Source: llm_adapter@2026-06-13

Confidence: high

Bugfix Medium

wait for in‑memory SQLite schema initialization

wait for in‑memory SQLite schema initialization

Source: llm_adapter@2026-06-13

Confidence: high

Full changelog

17.5.0 (2026-06-12)

Features

  • agent: add a subagents toggle to assistant settings (#13733) (1e8c6ac)
  • agents: improve agent product knowledge (#13705) (67ebb3a)
  • ui: pick a time range from a calendar in the time range selector (#13713) (94b8fd0)

Bug Fixes

  • add Anthropic computer-use beta header (#13242) (3d33799)
  • agent: focus PXI input on open (#13653) (d3d6e40)
  • deps: update arize-phoenix-client to 2.8.0 (#13676) (f93b1d4)
  • deps: update arize-phoenix-evals to 3.1.0 (#13011) (50f6227)
  • don't mask graphql syntax and validation errors (#13730) (e4f5396)
  • expire prompt tool diffs on provider change (#13641) (d5ba51b)
  • js: run package docs cleanup in postpublish instead of postpack (#13715) (9ef860a)
  • ui: use design system tooltip for project view mode toggle (#13724) (66fcfc0)
  • wait for in-memory sqlite schema initialization (#13637) (6fd5a8c)

Documentation

  • cookbook: dedupe + reframe custom LLM evaluator, fix create_dataset CSV bug (#13694) (174f4de)
  • June 10-11 release notes; allow PXI in release-notes skill (#13710) (4294d9d)
  • tracing: OpenInference best practices cookbook and concepts (#13538) (4a52ae2)
  • tutorials: level up trace-level evaluation cookbook (#13693) (fb6e2e6)

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]