Examples

Explore the capabilities of Phoenix with notebooks

Tutorials

LLM Ops - Tracing, Evaluation, and Analysis

Evaluation - Evaluating a RAG pipeline

Agents - Experimenting with Agents

Embedding Analysis - Search and Retrieval

Application Examples

Example full-stack applications instrumented using OpenInference and observed via phoenix server instances.

Python

LlamaIndex, Fast API Instrumentation

Javascript

OpenAI, Express, Next.JS

DSPy

DSPy, Fast API Instrumentation

TypeScript Chatbot with User Feedback

OpenAI, Express, Next.JS

Python Chatbot with User Feedback

OpenAI, Streamlit

Agent Demos

RAG Agent Chatbot

Code Generator Agent

Computer Use Agent

LLM Traces

Trace through the execution of your LLM application to understand its internal structure and to troubleshoot issues with retrieval, tool execution, LLM calls, and more.

Title	Topics	Links
Tracing an OpenAI App	Tracing Basics
Tracing a LlamaIndex App	Tracing Basics
Retrieval Example with Evaluations: Fast UI Viz	Evaluations Retrieval
Tracing and Evaluating a LlamaIndex + OpenAI RAG Application	LlamaIndex OpenAI retrieval-augmented generation
Tracing and Evaluating a LlamaIndex OpenAI Agent	LlamaIndex OpenAI agents function calling
Tracing and Evaluating a Structured Data Extraction Application with OpenAI Function Calling	OpenAI structured data extraction function calling
Tracing and Evaluating a LangChain + OpenAI RAG Application	LangChain OpenAI retrieval-augmented generation
Tracing and Evaluating a LangChain Agent	LangChain OpenAI agents function calling
Tracing and Evaluation a DSPy Application	LangChain Google PaLM retrieval-augmented generation
Tracing a LlamaIndex app with Sessions	LlamaIndex Tracing Sessions
Tracing an OpenAI app with Sessions	OpenAI Tracing Sessions
Tracing an OpenAI app with Sessions (JS/TS)	OpenAI Tracing Sessions

Datasets and Experiments

Iteratively improve your LLM task by building datasets, running experiments, and evaluating performance using code and LLM-as-a-judge.

Title	Topics	Links
Quickstart: Datasets and Experiments	datasets experiments
Text2SQL	SQL generation
Prompt Template Iteration for a Summarization Service	summarization
Answer Relevancy and Context Relevancy Evaluation	RAG
Guideline Eval	RAG
Pairwise Eval	pairwise eval
LlamaIndex RAG with Reranker	LlamaIndex RAG re-rankers
LangChain Email Extraction	LangChain

LLM Evals

Leverage the power of large language models to evaluate your generative model or application for hallucinations, toxicity, relevance of retrieved documents, and more.

Title	Topics	Links
Evaluating Hallucinations	hallucinations
Evaluating Toxicity	toxicity
Evaluating Relevance of Retrieved Documents	document relevance
Evaluating Question-Answering	question-answering
Evaluating Summarization	summarization
Evaluating Code Readability	code readability

Retrieval-Augmented Generation Analysis

Visualize your generative application's retrieval process to surface failed retrievals and to find topics not addressed by your knowledge base.

Title	Topics	Links
Evaluating and Improving Search and Retrieval Applications	LlamaIndex retrieval-augmented generation
Evaluating and Improving Search and Retrieval Applications	LlamaIndex Milvus retrieval-augmented generation

Embedding Analysis

Explore lower-dimensional representations of your embedding data to identify clusters of high-drift and performance degradation.

Title	Topics	Links
Active Learning for a Drifting Image Classification Model	image classification fine-tuning
Root-Cause Analysis for a Drifting Sentiment Classification Model	NLP sentiment classification
Troubleshooting an LLM Summarization Task	summarization
Collect Chats with GPT	LLMs
Find Clusters, Export, and Explore with GPT	LLMs exploratory data analysis

Structured Data Analysis

Statistically analyze your structured data to perform A/B analysis, temporal drift analysis, and more.

Title	Topics	Links
Detecting Fraud with Tabular Embeddings	tabular data anomaly detection