arize-phoenix-v16.0.0
Breaking Changes📦 arize-phoenixView on GitHub →
⚠ 1 breaking✨ 3 features🐛 1 fixes🔧 1 symbols
Summary
Version 16.0.0 introduces powerful Sandboxed Code Evaluators for flexible evaluation strategy composition and enables native web search/fetch for agents. This release contains breaking changes requiring migration.
⚠️ Breaking Changes
- Sandboxing and Code Evaluators were introduced, which likely changes how evaluation strategies are defined and executed. Users should consult MIGRATION.md for specific required changes.
Migration Steps
- Review the MIGRATION.md file for detailed instructions regarding the changes related to Sandboxing and Code Evaluators.
✨ New Features
- Introduction of Sandboxed Code Evaluators, allowing users to compose evaluation strategies by writing Python or TypeScript evaluate() functions directly in the Phoenix UI for server-side execution.
- Code Evaluators support composite scoring, embedding-based evaluation, and LLM juries.
- Agents now enable provider native web search/fetch when available.
🐛 Bug Fixes
- Fixed an issue preventing broken tool groups in agents.