@arizeai/phoenix-mcp

--- description: Available in Phoenix 8.9+ --- # 03.07.2025: New Prompt Playground, Evals, and Integration Support 🦾 {% embed url="https://storage.googleapis.com/arize-phoenix-assets/assets/gifs/added_support_release_notes.gif" %} New update overview: * **Prompt Playground**: Now supports [GPT-4.5 ](https://github.com/Arize-ai/phoenix/issues/6629)& Anthropic Sonnet 3.7 and Thinking Budgets * **Instrumentation**: [SmolagentsInstrumentor](broken-reference) to trace smolagents by Hugging Face * **Evals**: o3 support, Audio & Multi-Modal Evaluations * **Integrations**: Phoenix now supports [LiteLLM Proxy](broken-reference) & [Cleanlabs evals](broken-reference) ### Improvements and Bug Fixes 🐛 * [**Admin**](https://github.com/Arize-ai/phoenix/issues/6722)**:** Show percent used of DB * [**Configuration**](https://github.com/Arize-ai/phoenix/issues/6664): Add environment variable for allocated DB storage capacity * [**Tracing**](https://github.com/Arize-ai/phoenix/pull/6681): Delete selected traces * [**Tracing**](https://github.com/Arize-ai/phoenix/pull/6665): Make trace tree more readable on smaller sizes * [**Experiments**](https://github.com/Arize-ai/phoenix/pull/6708): Ensure type is correct on `run_experiment` * [**Experiments**](https://github.com/Arize-ai/phoenix/pull/6642): Allow experiment run JSON downloads * [**Python Client**](https://github.com/Arize-ai/phoenix/issues/6659): Add anthropic thinking config param * [**Components**](https://github.com/Arize-ai/phoenix/pull/6679): Add ToggleButton