Arize-ai/phoenix arize-phoenix-evals-v2.8.0

arize-phoenix-evals: v2.8.0

on GitHub

2.8.0 (2026-01-08)

Features

• correctness evaluator (#10829) (da13ad5)
• Enable sync- and async- client kwargs for LLM constructor (#10808) (1442074)
• evals: add built-in tool selection correctness metric (#10744) (9e816c9)

Bug Fixes

• prompts: remove newlines from built-in prompts (#10736) (edf35f0)