phoenix

varize-phoenix-v17.30.0 scope: arize-phoenix Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 13d Tracing

View tool

✓ 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, 17.30.0, and 2026-07-14 across a mixed release.

Changes in this release

Type	Severity	Summary	CVE
Feature
Feature	Low	agents: force tracing with debug env variable agents: force tracing with debug env variable Source: llm_adapter@2026-07-14 Confidence: high	—
Feature	Low	app: support batch annotation config management app: support batch annotation config management Source: llm_adapter@2026-07-14 Confidence: high	—
Feature	Low	playground: show output error count playground: show output error count Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix
Bugfix	Medium	add NULLS LAST to projects page end‑time sort order add NULLS LAST to projects page end‑time sort order Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix	Medium	agents: exclude browser clock from system prompt to preserve caching agents: exclude browser clock from system prompt to preserve caching Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix	Medium	agents: use Responses API for PXI OpenAI models agents: use Responses API for PXI OpenAI models Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix	Medium	pxi: export agent telemetry from pytest evaluation runs pxi: export agent telemetry from pytest evaluation runs Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix	Medium	resolve PostgreSQL schema for database usage stats collection resolve PostgreSQL schema for database usage stats collection Source: llm_adapter@2026-07-14 Confidence: high	—
Bugfix	Low	ui: allow empty prompt descriptions in playground output header test ui: allow empty prompt descriptions in playground output header test Source: llm_adapter@2026-07-14 Confidence: high	—

Full changelog

17.30.0 (2026-07-14)

Features

agents: force tracing with debug env (#14358) (1dbf5ff)
app: support batch annotation config management (#14198) (12be3cc)
playground: show output error count (#14277) (8a75d67)

Bug Fixes

add nullslast to projects page end time sort (#14092) (36d0e6e)
agents: keep browser clock out of the system prompt to preserve prompt caching (#14362) (c5dbb20)
agents: use Responses API for PXI OpenAI models (#14365) (61fc1e4)
pxi: export agent telemetry from pytest eval runs (#14336) (e4b94b4)
resolve PostgreSQL schema for database usage stats (#14172) (c26fdfe)
theme provider in playground output header test (#14351) (33564d5)
ui: allow empty prompt descriptions (#14284) (c03ea6c)

View diff on GitHub

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

Share on X Share on Bluesky

Track phoenix

Get notified when new releases ship.

About phoenix

AI Observability & Evaluation

All releases →

Related context

Related tools

Earlier breaking changes

varize-phoenix-v19.0.0 GraphQL createUserApiKey and createSystemApiKey reject API-key callers.
varize-phoenix-v18.0.0 Changes session time-range filters to use interval-overlap semantics.
varize-phoenix-v17.0.0 Adds system settings for admin-managed assistant enablement and trace recording policy
varize-phoenix-v16.0.0 Sandboxing and Code Evaluators introduce breaking changes in Phoenix v16.0.0.
varize-phoenix-v15.7.0 Removes v1 /chat route and associated code