# Phoenix

> Ask these docs: GET /ask?q=<keywords or question> returns JSON { results: [{ href, title, heading, snippet, tab }] }; short keywords rank best. Append .md to any page href for its markdown.

## Documentation

- [What is Phoenix?](/docs/phoenix.md): AI Observability and Evaluation
- [Start Phoenix](/docs/phoenix/get-started.md): Run Phoenix locally, in a container, or on your own infrastructure, then send it your first trace.
- [Observe](/docs/phoenix/get-started/get-started-tracing.md): Trace a model call and send it to Phoenix, in Python or TypeScript.
- [Evaluate](/docs/phoenix/get-started/get-started-evaluations.md): Build a dataset, write an LLM judge, and compare two prompts as experiments, in Python or TypeScript.
- [Connect Your Coding Agent](/docs/phoenix/integrations/developer-tools/coding-agents.md): Connect your coding agent to Phoenix so it can instrument your app, read the traces that come back, and act on what it finds.
- [CLI](/docs/phoenix/integrations/developer-tools/cli.md): Let any coding agent query your Phoenix traces, datasets, experiments, and prompts from the terminal.
- [MCP Server](/docs/phoenix/integrations/mcp.md): Connect Claude Code, Cursor, VS Code, or any MCP client to your live Phoenix data.
- [Skills](/docs/phoenix/integrations/developer-tools/skills.md): Teach your coding agent Phoenix's conventions for the CLI, tracing, and evals, so it gets them right the first time.
- [PXI](/docs/phoenix/pxi.md): The AI engineering agent built into Phoenix. Hand it an investigation instead of running it yourself.
- [Advanced](/docs/phoenix/pxi-advanced.md): Skills, the permission model, the terminal client, slash commands, model providers, and configuration.
- [Tracing Tutorial](/docs/phoenix/tracing/tutorial.md): Build a fully observable AI agent from scratch - trace every operation, measure quality, and debug conversations
- [Your First Traces](/docs/phoenix/tracing/tutorial/your-first-traces.md): Build a support agent and trace every LLM call, tool execution, and RAG retrieval
- [Annotations and Evaluation](/docs/phoenix/tracing/tutorial/annotations-and-evaluations.md): Is your agent actually good? Use annotations and LLM-as-Judge to find out.
- [Sessions](/docs/phoenix/tracing/tutorial/sessions.md): Track multi-turn conversations as cohesive units in Phoenix
- [Overview](/docs/phoenix/tracing/llm-traces.md): Tracing the execution of LLM applications using Telemetry
- [Projects](/docs/phoenix/tracing/llm-traces/projects.md): Use projects to organize your LLM traces
- [Sessions](/docs/phoenix/tracing/llm-traces/sessions.md): Track and analyze multi-turn conversations
- [Annotations](/docs/phoenix/tracing/llm-traces/how-to-annotate-traces.md): In order to improve your LLM application iteratively, it's vital to collect feedback, annotate data during human review, as well as to establish an…
- [Metrics Dashboard](/docs/phoenix/tracing/llm-traces/metrics.md): Each Phoenix project comes with a pre-defined metrics dashboard
- [Overview](/docs/phoenix/tracing/how-to-tracing.md): Guides on how to use traces
- [Overview](/docs/phoenix/tracing/how-to-tracing/setup-tracing.md): Configure tracing for your application
- [Setup OTEL](/docs/phoenix/tracing/how-to-tracing/setup-tracing/setup-using-phoenix-otel.md): Configure OpenTelemetry tracing with Phoenix-aware defaults
- [Using Tracing Helpers](/docs/phoenix/tracing/how-to-tracing/setup-tracing/instrument.md): OpenInference packages provide helpful abstractions to make manual instrumentation of agents simpler.
- [Setup Projects](/docs/phoenix/tracing/how-to-tracing/setup-tracing/setup-projects.md): Log to a specific project Phoenix uses projects to group traces. If left unspecified, all traces are sent to a default project.
- [Setup Sessions](/docs/phoenix/tracing/how-to-tracing/setup-tracing/setup-sessions.md): How to track sessions across multiple traces
- [Overview](/docs/phoenix/tracing/how-to-tracing/add-metadata.md): Tracing can be augmented and customized by adding Metadata. Metadata includes your own custom attributes, user ids, session ids, prompt templates, and more.
- [Add Attributes, Metadata, Users](/docs/phoenix/tracing/how-to-tracing/add-metadata/customize-spans.md): Using context to customize spans In order to customize spans that are created via auto-instrumentation, The Otel Context can be used to set span…
- [Instrument Prompt Templates and Prompt Variables](/docs/phoenix/tracing/how-to-tracing/add-metadata/instrumenting-prompt-templates-and-prompt-variables.md): Instrumenting prompt templates and variables allows you to track and visualize prompt changes.
- [Overview](/docs/phoenix/tracing/how-to-tracing/feedback-and-annotations.md): Annotating traces is a crucial aspect of evaluating and improving your LLM-based applications.
- [Annotating in the UI](/docs/phoenix/tracing/how-to-tracing/feedback-and-annotations/annotating-in-the-ui.md): How to annotate traces in the UI for analysis and dataset curation
- [Annotating via the Client](/docs/phoenix/tracing/how-to-tracing/feedback-and-annotations/capture-feedback.md): Use the phoenix client to capture end-user feedback
- [Annotating Auto-Instrumented Spans](/docs/phoenix/tracing/how-to-tracing/feedback-and-annotations/annotating-auto-instrumented-spans.md): Use the capturespancontext context manager to annotate auto-instrumented spans
- [Running Evals on Traces](/docs/phoenix/tracing/how-to-tracing/feedback-and-annotations/evaluating-phoenix-traces.md): How to use an LLM judge to label and score your application
- [Log Evaluation Results](/docs/phoenix/tracing/how-to-tracing/feedback-and-annotations/llm-evaluations.md): Evaluations, which can be considered a form of automated annotation, are logged as annotations inside of Phoenix.
- [Overview](/docs/phoenix/tracing/how-to-tracing/importing-and-exporting-traces.md): Learn how to load a file of traces into Phoenix Upload agent trajectories from Harbor, Claude Code, OpenHands, and other ATIF-compatible frameworks…
- [Import Existing Traces](/docs/phoenix/tracing/how-to-tracing/importing-and-exporting-traces/importing-existing-traces.md): Phoenix supports loading data that contains OpenInference traces. This allows you to load an existing dataframe of traces into your Phoenix instance.
- [Import ATIF trajectories](/docs/phoenix/tracing/how-to-tracing/importing-and-exporting-traces/importing-atif-trajectories.md): Convert saved ATIF agent trajectories into Phoenix traces.
- [Export Data & Query Spans](/docs/phoenix/tracing/how-to-tracing/importing-and-exporting-traces/extract-data-from-spans.md): Various options to help you get data out of Phoenix
- [Exporting Annotated Spans](/docs/phoenix/tracing/how-to-tracing/importing-and-exporting-traces/exporting-annotated-spans.md): Span annotations can be an extremely valuable basis for improving your application. The Phoenix client provides useful ways to pull down spans and their associated annotations.
- [Retrieve Traces via CLI](/docs/phoenix/tracing/how-to-tracing/importing-and-exporting-traces/retrieve-traces-via-cli.md): Fetch trace data from Phoenix using the command-line interface for debugging, analysis, and integration with AI coding assistants.
- [Cost Tracking](/docs/phoenix/tracing/how-to-tracing/cost-tracking.md): Phoenix allows you to track token-based costs for LLM runs automatically.
- [Filter Expressions](/docs/phoenix/tracing/how-to-tracing/filter-expressions.md): Filter spans, traces, and sessions with Python-style filter expressions, in the UI and over the API.
- [Overview](/docs/phoenix/tracing/how-to-tracing/advanced.md): Learn how to block PII from logging to Phoenix Learn how to selectively block or turn off tracing Learn how to send only certain spans to Phoenix…
- [Mask Span Attributes](/docs/phoenix/tracing/how-to-tracing/advanced/masking-span-attributes.md): In some situations, you may need to modify the observability level of your tracing. For instance, you may want to keep sensitive information from being logged for security reasons, or you may want to limit the size of the base64 encoded images logged to reduced payload size.
- [Suppress Tracing](/docs/phoenix/tracing/how-to-tracing/advanced/suppress-tracing.md): How to turn off tracing Tracing can be paused temporarily or disabled permanently.
- [Filter Spans to Export](/docs/phoenix/tracing/how-to-tracing/advanced/modifying-spans.md): Sometimes while instrumenting your application, you may want to filter out or modify certain spans from being sent to Phoenix.
- [Capture Multimodal Traces](/docs/phoenix/tracing/how-to-tracing/advanced/multimodal-tracing.md): Phoenix supports displaying images that are included in LLM traces.
- [Run Evals With Built-In Eval Templates](/docs/phoenix/evaluation/tutorials/run-evals-with-built-in-evals.md): This guide is part of a sequence: it starts with built-in eval templates, then moves to customizing the judge model, then to defining your own…
- [Customize Your LLM Endpoint](/docs/phoenix/evaluation/tutorials/customize-your-llm-endpoint.md): Phoenix Evals gives you flexibility in how you configure the model that acts as a judge.
- [Customize Your Evaluation Template](/docs/phoenix/evaluation/tutorials/customize-eval-template.md): Built-in eval templates cover many common evaluation patterns, but they can’t capture every application-specific requirement.
- [Overview](/docs/phoenix/evaluation/llm-evals.md): Evaluations measure the quality of your AI application's outputs — whether responses are accurate, grounded, safe, or relevant to the user's intent.
- [LLM as a Judge](/docs/phoenix/evaluation/concepts-evals/llm-as-a-judge.md): Evaluating tasks performed by LLMs can be difficult due to their complexity and the diverse criteria involved. Traditional methods like rule-based assessment or similarity metrics (e.g., ROUGE, BLEU) often fall short when applied to the nuanced and varied outputs of LLMs.
- [Overview](/docs/phoenix/evaluation/how-to-evals.md): The Phoenix Evals SDK provides composable building blocks for writing and running evaluations in Python or TypeScript.
- [Custom LLM Evaluators](/docs/phoenix/evaluation/how-to-evals/custom-llm-evaluators.md): Building Custom Evaluators While pre-built evals offer convenience, the best evals are ones you custom build for your specific use case.
- [Configuring the LLM](/docs/phoenix/evaluation/how-to-evals/configuring-the-llm.md): LLM Evaluators require an LLM in order to score an evaluation input. Phoenix evals are provider agnostic and work with virtually any foundation model.
- [Prompt Templates](/docs/phoenix/evaluation/how-to-evals/prompt-formats.md): Phoenix evaluators support multiple prompt formats, all compatible with supported models and providers.
- [Code Evaluators](/docs/phoenix/evaluation/how-to-evals/code-evaluators.md): Evaluations do not all require LLMs, and often it's useful to create Evaluators that perform basic checks or calculations on datasets that, in…
- [Code Evaluator Output Shapes](/docs/phoenix/evaluation/how-to-evals/code-evaluator-output-shapes.md): This page documents every return shape a code evaluator can produce and how Phoenix maps each one to an EvaluationResult with label, score, and…
- [Batch Evaluations](/docs/phoenix/evaluation/how-to-evals/batch-evaluations.md): Dataframe Evaluation Methods (Python only) evaluatedataframe for synchronous dataframe evaluations asyncevaluatedataframe an asynchronous version for…
- [Using Evals with Phoenix](/docs/phoenix/evaluation/how-to-evals/using-evals-with-phoenix.md): The evals library is designed to work independently — you can run evaluations without any other part of Phoenix. That said, it integrates naturally with tracing, datasets, and experiments when you need it. For more information about how to use the evals library with other Phoenix features, reference these guides:
- [Overview](/docs/phoenix/evaluation/pre-built-metrics.md): Ready-to-use evaluation metrics for measuring LLM application quality
- [Faithfulness](/docs/phoenix/evaluation/pre-built-metrics/faithfulness.md): Evaluate whether LLM responses are faithful to the provided context.
- [Hallucination](/docs/phoenix/evaluation/pre-built-metrics/hallucination.md): Detect whether an assistant response contains claims unsupported by the conversation.
- [Completeness](/docs/phoenix/evaluation/pre-built-metrics/completeness.md): Assess whether every active user request in a conversation was actually completed.
- [Conciseness](/docs/phoenix/evaluation/pre-built-metrics/conciseness.md): Evaluate whether LLM responses are concise and free of unnecessary content.
- [Correctness](/docs/phoenix/evaluation/pre-built-metrics/correctness.md): Evaluate whether LLM responses are generally correct and complete.
- [Retrieval Relevance](/docs/phoenix/evaluation/pre-built-metrics/retrieval-relevance.md): Evaluate whether externally retrieved information is relevant to the request it was serving.
- [Tool Selection](/docs/phoenix/evaluation/pre-built-metrics/tool-selection.md): Evaluate whether LLMs select the correct tools for given tasks
- [Tool Invocation](/docs/phoenix/evaluation/pre-built-metrics/tool-invocation.md): Evaluate whether LLM tool calls have correct arguments and formatting
- [Tool Response Handling](/docs/phoenix/evaluation/pre-built-metrics/tool-response-handling.md): Evaluate whether AI agents correctly process tool results, including error handling, data extraction, and safe information disclosure
- [Refusal](/docs/phoenix/evaluation/pre-built-metrics/refusal.md): Detect when an LLM refuses or declines to answer a user query.
- [Toxicity](/docs/phoenix/evaluation/pre-built-metrics/toxicity.md): Detect whether text is toxic — hateful, demeaning, abusive, or threatening.
- [User Friction](/docs/phoenix/evaluation/pre-built-metrics/user-friction.md): Detect when a user expresses friction with an assistant's preceding behavior.
- [PII Detection](/docs/phoenix/evaluation/pre-built-metrics/pii-detection.md): Detect personally identifiable information in a conversation record.
- [Exact Match](/docs/phoenix/evaluation/pre-built-metrics/exact-match.md): Evaluate if output exactly matches expected value
- [Matches Regex](/docs/phoenix/evaluation/pre-built-metrics/matches-regex.md): Evaluate if output matches a regular expression pattern
- [Precision / Recall / F-Score](/docs/phoenix/evaluation/pre-built-metrics/precision-recall-fscore.md): Compute precision, recall, and F-beta scores for classification tasks
- [Overview](/docs/phoenix/evaluation/server-evals/overview.md): Attach evaluators directly to a dataset and run them automatically on every experiment.
- [LLM Evaluators](/docs/phoenix/evaluation/server-evals/llm-evaluators.md): LLM-as-a-judge evaluators that use Phoenix-managed prompts to label and score experiment outputs with full version control and traceability.
- [Code Evaluators](/docs/phoenix/evaluation/server-evals/code-evaluators.md): Write Python or TypeScript evaluators directly in Phoenix and run them in a managed sandbox — no SDK, local runtime, or deploy step required.
- [JSON Distance](/docs/phoenix/evaluation/server-evals/code-evaluators/json-distance.md): Count the structural differences between an output JSON and a reference from a golden dataset.
- [Regex Match](/docs/phoenix/evaluation/server-evals/code-evaluators/regex-match.md): Pass when the output matches a regular expression pattern.
- [Embedding Distance](/docs/phoenix/evaluation/server-evals/code-evaluators/embedding-distance.md): Score semantic similarity between two strings using an embeddings API.
- [scikit-learn](/docs/phoenix/evaluation/server-evals/code-evaluators/scikit-learn.md): Score text similarity offline with scikit-learn's HashingVectorizer and cosine similarity — no API calls.
- [Pairwise Evaluator](/docs/phoenix/evaluation/server-evals/code-evaluators/pairwise.md): Blind LLM-judge head-to-head comparison of the model output against a reference answer.
- [Composite Evaluator](/docs/phoenix/evaluation/server-evals/code-evaluators/composite.md): Blend multiple sub-scores into a single weighted-average score with a per-axis breakdown.
- [LLM Jury](/docs/phoenix/evaluation/server-evals/code-evaluators/llm-jury.md): Poll multiple LLMs as judges and combine their verdicts with a weighted average to reduce single-model bias.
- [Input Mapping](/docs/phoenix/evaluation/server-evals/input-mapping.md): Control how dataset and task output values flow into evaluator inputs using JSONPath expressions and literal values.
- [Overview](/docs/phoenix/evaluation/server-evals/pre-built-metrics.md): Ready-to-use server-side evaluators for common evaluation tasks.
- [Contains](/docs/phoenix/evaluation/server-evals/pre-built-metrics/contains.md): Check whether a text contains one or more specified words.
- [Exact Match](/docs/phoenix/evaluation/server-evals/pre-built-metrics/exact-match.md): Check whether two strings are identical.
- [Regex Match](/docs/phoenix/evaluation/server-evals/pre-built-metrics/regex.md): Check whether a text matches a regular expression pattern.
- [Levenshtein Distance](/docs/phoenix/evaluation/server-evals/pre-built-metrics/levenshtein-distance.md): Measure the edit distance between two strings.
- [JSON Distance](/docs/phoenix/evaluation/server-evals/pre-built-metrics/json-distance.md): Measure the number of structural differences between two JSON values.
- [Correctness](/docs/phoenix/evaluation/server-evals/pre-built-metrics/correctness.md): Evaluate whether LLM responses are generally correct and complete using a Phoenix-managed judge model.
- [Tool Selection](/docs/phoenix/evaluation/server-evals/pre-built-metrics/tool-selection.md): Evaluate whether LLMs select the correct tools for given tasks using a Phoenix-managed judge model.
- [Tool Invocation](/docs/phoenix/evaluation/server-evals/pre-built-metrics/tool-invocation.md): Evaluate whether LLM tool calls have correct arguments and formatting using a Phoenix-managed judge model.
- [Harbor](/docs/phoenix/integrations/evaluation-integrations/harbor.md): Record Harbor agent evaluations as Phoenix datasets, experiments, scores, and ATIF traces.
- [pytest](/docs/phoenix/evaluation/integrations/pytest.md): Write LLM evaluations as ordinary pytest tests that run in CI and record results to Phoenix.
- [Vitest / Jest](/docs/phoenix/evaluation/integrations/vitest-jest.md): Write LLM evaluations as Vitest or Jest tests that record results to Phoenix and gate CI.
- [Defining the Dataset That Powers Your Experiments](/docs/phoenix/datasets-and-experiments/tutorial/defining-the-dataset.md): A dataset is the foundation for systematic evaluation and iterative improvement in your AI workflow.
- [Run Experiments with Code Evals](/docs/phoenix/datasets-and-experiments/tutorial/run-experiments-with-code-evals.md): Learn how to define and run experiments to systematically evaluate your AI application using code-based evaluators and ground truth data.
- [Run Experiments with LLM as a Judge](/docs/phoenix/datasets-and-experiments/tutorial/run-experiments-with-llm-judge.md): Learn how to define and run experiments to systematically evaluate your AI application using LLM as a Judge evaluators for subjective quality assessments.
- [Iterating with Experiments in Your Workflow](/docs/phoenix/datasets-and-experiments/tutorial/iteration-workflow-experiments.md): Learn how to use experiments to systematically validate changes to your AI application and compare different versions over time.
- [Overview: Datasets & Experiments](/docs/phoenix/datasets-and-experiments/overview-datasets.md): The velocity of AI application development is bottlenecked by quality evaluations because AI engineers are often faced with hard tradeoffs: which…
- [Overview](/docs/phoenix/datasets-and-experiments/how-to-datasets.md): Datasets are critical assets for building robust prompts, evals, fine-tuning,
- [Creating Datasets](/docs/phoenix/datasets-and-experiments/how-to-datasets/creating-datasets.md): From CSV When manually creating a dataset (let's say collecting hypothetical questions and answers), the easiest way to start is by using a…
- [Updating Datasets](/docs/phoenix/datasets-and-experiments/how-to-datasets/updating-datasets.md): Update existing dataset examples using stable IDs with arize-phoenix-client
- [Exporting Datasets](/docs/phoenix/datasets-and-experiments/how-to-datasets/exporting-datasets.md): Exporting to CSV Want to just use the contents of your dataset in another context?
- [Linking Examples to Their Source Span (REST)](/docs/phoenix/datasets-and-experiments/how-to-datasets/linking-examples-to-spans.md): Link dataset examples back to the spans and traces they came from over Phoenix's REST API — via example metadata or the source span link — then log evaluations onto those spans.
- [Overview](/docs/phoenix/datasets-and-experiments/how-to-experiments.md): Running Experiments Phoenix supports two workflows for experiments: a UI-driven flow in the Playground and a programmatic SDK flow.
- [Run Experiments](/docs/phoenix/datasets-and-experiments/how-to-experiments/running-experiments.md): Run experiments from the Playground or programmatically with the SDK.
- [Evaluate a Prompt Change in Code](/docs/phoenix/datasets-and-experiments/how-to-experiments/evaluate-a-prompt-change.md): Build the dataset, task, evaluator, and experiment with the SDK, then compare a baseline prompt against an edited one.
- [Run Experiments in the Background](/docs/phoenix/datasets-and-experiments/how-to-experiments/running-experiments-in-background.md): Start experiments from the Playground that keep running after you close the browser, survive server restarts, and can be stopped or resumed at any time.
- [Using Evaluators](/docs/phoenix/datasets-and-experiments/how-to-experiments/using-evaluators.md): Evaluators are a way of validating that your AI task is running as expected.
- [Dataset Evaluators](/docs/phoenix/datasets-and-experiments/how-to-experiments/how-to-dataset-evaluators.md): Attach evaluators to datasets for automatic scoring during experiments.
- [Repetitions](/docs/phoenix/datasets-and-experiments/how-to-experiments/repetitions.md): How to leverage repetitions to get an understanding of indeterminate LLM outputs
- [Splits](/docs/phoenix/datasets-and-experiments/how-to-experiments/splits.md): How to run experiments over select splits of your dataset for targeted experimentation
- [Eval CI with pytest](/docs/phoenix/datasets-and-experiments/how-to-experiments/eval-ci-with-pytest.md): Run LLM evals as pytest tests and record them as Phoenix experiments
- [Overview](/docs/phoenix/prompt-engineering/tutorial.md): Phoenix Prompts Tutorial
- [Identify & Edit Prompts](/docs/phoenix/prompt-engineering/tutorial/identify-and-edit-prompts.md): Fix and store bad prompts from your spans
- [Test Prompts at Scale](/docs/phoenix/prompt-engineering/tutorial/test-prompts-at-scale.md): Measure and Edit Prompts at Scale
- [Compare Prompt Versions](/docs/phoenix/prompt-engineering/tutorial/compare-prompt-versions.md): Build New Prompt Versions and Compare
- [Optimize Prompts Automatically](/docs/phoenix/prompt-engineering/tutorial/optimize-prompts-automatically.md): Automatically Optimize Prompts with Prompt Learning
- [Overview](/docs/phoenix/prompt-engineering/overview-prompts.md): Prompt management allows you to create, store, and modify prompts for interacting with LLMs.
- [Prompt Management](/docs/phoenix/prompt-engineering/overview-prompts/prompt-management.md): Version and track changes made to prompt templates
- [Prompt Playground](/docs/phoenix/prompt-engineering/overview-prompts/prompt-playground.md): Phoenix's Prompt Playground makes the process of iterating and testing prompts quick and easy.
- [Span Replay](/docs/phoenix/prompt-engineering/overview-prompts/span-replay.md): Replay LLM spans traced in your application directly in the playground
- [Prompts in Code](/docs/phoenix/prompt-engineering/overview-prompts/prompts-in-code.md): Pull and push prompt changes via Phoenix's Python and TypeScript Clients
- [Overview](/docs/phoenix/prompt-engineering/how-to-prompts.md): Guides on how to do prompt engineering with Phoenix
- [Configure AI Providers](/docs/phoenix/prompt-engineering/how-to-prompts/configure-ai-providers.md): Phoenix natively integrates with OpenAI, Azure OpenAI, Anthropic, and Google AI Studio (gemini) to make it easy to test changes to your prompts. In addition to the above, since many AI providers (deepseek, ollama) can be used directly with the OpenAI client, you can talk to any OpenAI compatible LLM provider.
- [Using the Playground](/docs/phoenix/prompt-engineering/how-to-prompts/using-the-playground.md): General guidelines on how to use Phoenix's prompt playground
- [Create a prompt](/docs/phoenix/prompt-engineering/how-to-prompts/create-a-prompt.md): Store and track prompt versions in Phoenix
- [Test a prompt](/docs/phoenix/prompt-engineering/how-to-prompts/test-a-prompt.md): Testing your prompts before you ship them is vital to deploying reliable AI applications
- [Tag a prompt](/docs/phoenix/prompt-engineering/how-to-prompts/tag-a-prompt.md): How to deploy prompts to different environments safely
- [Using a prompt](/docs/phoenix/prompt-engineering/how-to-prompts/using-a-prompt.md): Once you have tagged a version of a prompt as ready (e.x. "staging") you can pull a prompt into your code base and use it to prompt an LLM.
- [Use provider tools](/docs/phoenix/prompt-engineering/how-to-prompts/use-provider-tools.md): Use built-in, vendor-specific tools from OpenAI, Anthropic, Google, and Amazon Bedrock with Phoenix prompts
- [Access Control (RBAC)](/docs/phoenix/settings/access-control-rbac.md): The role-based access control (RBAC) in Phoenix is based on the following user roles: admin - full control to the system, can administer users,…
- [API Keys](/docs/phoenix/settings/api-keys.md): Phoenix offers three types of authentication credentials: System API Keys, User API Keys, and the Admin Secret System API Keys System keys…
- [Secrets](/docs/phoenix/settings/secrets.md): Store and manage encrypted API keys and credentials in Phoenix.
- [Data Retention](/docs/phoenix/settings/data-retention.md): By default Phoenix will store and preserve all your data and data retention is entirely under your control.
- [Custom AI Providers](/docs/phoenix/settings/custom-ai-providers.md): Configure server-managed AI providers for shared use across the Phoenix platform.
- [Sandboxes](/docs/phoenix/settings/sandboxes.md): Configure the sandbox backends that execute code evaluators and other server-side code workloads.
- [WebAssembly](/docs/phoenix/settings/sandboxes/wasm.md): Run simple Python code evaluators locally inside the Phoenix server.
- [Deno](/docs/phoenix/settings/sandboxes/deno.md): Run simple TypeScript code evaluators locally inside the Phoenix server.
- [E2B](/docs/phoenix/settings/sandboxes/e2b.md): Run code evaluators in E2B's hosted micro-VM sandboxes.
- [Daytona](/docs/phoenix/settings/sandboxes/daytona.md): Run code evaluators in Daytona's managed development sandboxes.
- [Vercel Sandbox](/docs/phoenix/settings/sandboxes/vercel.md): Run code evaluators in ephemeral compute on Vercel's infrastructure.
- [Modal](/docs/phoenix/settings/sandboxes/modal.md): Run code evaluators in Modal's serverless container platform.
- [User Guide](/docs/phoenix/user-guide.md): Phoenix is a comprehensive platform designed to enable observability across every layer of an LLM-based system, empowering teams to build, optimize, and maintain high-quality applications and agents efficiently.
- [Production Guide](/docs/phoenix/production-guide.md): Moving your application to production: steps for reliability and scale
- [Environments](/docs/phoenix/environments.md): The Phoenix app can be run in various environments such as Colab and SageMaker notebooks, as well as be served via the terminal or a docker container.
- [What are Traces](/docs/phoenix/tracing/concepts-tracing/what-are-traces.md): Spans A span represents a unit of work or operation (think a span of time).
- [How Tracing Works](/docs/phoenix/tracing/concepts-tracing/how-tracing-works.md): Instrumentation In order for an application to emit traces for analysis, the application must be instrumented.
- [Annotations Concepts](/docs/phoenix/tracing/concepts-tracing/annotations-concepts.md): Annotation Types Depending on what you want to do with your annotations, you may want to configure a rubric for what your annotation represents - e.g.
- [Translating Semantic Conventions](/docs/phoenix/tracing/concepts-tracing/translating-conventions.md): Convert traces from OpenLIT, OpenLLMetry, and OpenTelemetry GenAI to OpenInference format so they appear in Phoenix
- [Overview](/docs/phoenix/tracing/concepts-tracing/otel-openinference/overview.md): How Phoenix tracing is built on OpenTelemetry and OpenInference — the two open standards that make AI observability vendor-agnostic and portable.
- [Signals, spans, traces, and sessions](/docs/phoenix/tracing/concepts-tracing/otel-openinference/signals.md): The data shapes OpenTelemetry produces — signals, spans, traces, and sessions — and how they fit together into the trace tree Phoenix visualizes.
- [Resource](/docs/phoenix/tracing/concepts-tracing/otel-openinference/resource.md): The immutable metadata that describes who is producing telemetry — service name, environment, host, and Phoenix attributes that route spans to your project.
- [Exporter and OTLP](/docs/phoenix/tracing/concepts-tracing/otel-openinference/exporter.md): How OpenTelemetry ships spans out of your application — exporters, the OTLP protocol, and the choice between gRPC and HTTP transport.
- [Span processor](/docs/phoenix/tracing/concepts-tracing/otel-openinference/span-processor.md): The OpenTelemetry component that intercepts spans at the start and end of their lifecycle — how BatchSpanProcessor and SimpleSpanProcessor differ, and how to tune them.
- [Tracer provider and tracer](/docs/phoenix/tracing/concepts-tracing/otel-openinference/tracer-provider.md): The central configuration object for OpenTelemetry tracing — what the Tracer Provider owns, how to configure it, and the Lambda and Node.js shutdown gotchas to know about.
- [phoenix.otel helpers](/docs/phoenix/tracing/concepts-tracing/otel-openinference/phoenix-otel-helpers.md): The arize-phoenix-otel SDK collapses raw OpenTelemetry setup into a single register() call, with auto-instrumentation discovery and Phoenix-aware defaults.
- [Semantic conventions](/docs/phoenix/tracing/concepts-tracing/otel-openinference/semantic-conventions.md): Standardized attribute names that make AI trace data portable across tools, languages, and platforms — and how OpenInference compares to the still-evolving GenAI conventions.
- [Span kinds](/docs/phoenix/tracing/concepts-tracing/otel-openinference/span-kinds.md): The twelve OpenInference span kinds — LLM, Tool, Agent, Chain, Retriever, and the rest — with the canonical attributes Phoenix expects on each.
- [Instrumentation approaches](/docs/phoenix/tracing/concepts-tracing/otel-openinference/instrumentation-approaches.md): Three ways to emit spans — auto-instrumentation, manual instrumentation, and hybrid — with trade-offs, mechanics, and the custom span processor pattern.
- [Context managers](/docs/phoenix/tracing/concepts-tracing/otel-openinference/context-managers.md): Six context managers from openinference.instrumentation that attach session IDs, user IDs, metadata, tags, and prompt templates to every span in scope.
- [Context propagation](/docs/phoenix/tracing/concepts-tracing/otel-openinference/context-propagation.md): How OpenTelemetry carries trace context across async boundaries, threads, and microservices — when propagation is automatic, when it's manual, and which propagators to use.
- [Sampling](/docs/phoenix/tracing/concepts-tracing/otel-openinference/sampling.md): Head and tail sampling — how to reduce telemetry volume without losing visibility, configure samplers via SDK or environment variables, and avoid the silent alwaysoff gotcha.
- [OpenTelemetry Collector](/docs/phoenix/tracing/concepts-tracing/otel-openinference/otel-collector.md): An optional intermediate service that receives spans, processes them, and exports them to a backend — with deployment models, use cases, and the most common configuration pitfalls.
- [FAQs: Tracing](/docs/phoenix/tracing/concepts-tracing/faqs-tracing.md): How to log traces To log traces, you must instrument your application either manually or automatically.
- [Prompts Concepts](/docs/phoenix/prompt-engineering/concepts-prompts/prompts-concepts.md): Prompt Prompts often times refer to the content of how you "prompt" a LLM, e.g. the "text" that you send to a model like OpenAI's gpt-4.
- [Context Engineering Basics](/docs/phoenix/prompt-engineering/concepts-prompts/context-engineering-basics.md): What Is Context Engineering?
- [Concepts: Datasets](/docs/phoenix/datasets-and-experiments/concepts-datasets.md): Datasets Datasets are integral to evaluation and experimentation.
- [Overview](/docs/phoenix/resources/frequently-asked-questions.md): For support, join the community Slack to ask questions and connect with other developers.
- [What is my Phoenix Endpoint?](/docs/phoenix/resources/frequently-asked-questions/what-is-my-phoenix-endpoint.md): There are two endpoints that matter in Phoenix: Application Endpoint: The endpoint your Phoenix instance is running on OTEL Tracing Endpoint: The…
- [Can I add other users to my Phoenix Instance?](/docs/phoenix/resources/frequently-asked-questions/can-i-add-other-users-to-my-phoenix-instance.md): Yes. Phoenix supports multiple users through authentication, with role-based access control, API keys, and OAuth2/SSO.
- [Can I use Azure OpenAI?](/docs/phoenix/resources/frequently-asked-questions/can-i-use-azure-openai.md): Yes, in fact this is probably the preferred way to interact with OpenAI if your enterprise requires data privacy.
- [Can I use Phoenix locally from a remote Jupyter instance?](/docs/phoenix/resources/frequently-asked-questions/can-i-use-phoenix-locally-from-a-remote-jupyter-instance.md): Yes, you can use either of the two methods below.
- [How can I configure the backend to send the data to the phoenix UI in another container?](/docs/phoenix/resources/frequently-asked-questions/how-can-i-configure-the-backend-to-send-the-data-to-the-phoenix-ui-in-another-container.md): If you are working on an API whose endpoints perform RAG, but would like the phoenix server not to be launched as another thread.
- [Can I run Phoenix on Sagemaker?](/docs/phoenix/resources/frequently-asked-questions/can-i-run-phoenix-on-sagemaker.md): With SageMaker notebooks, phoenix leverages the jupyter-server-proy to host the server under proxy/6006.Note, that phoenix will automatically try to detect that you are running in SageMaker but you can declare the notebook runtime via a parameter to launchapp or an environment variable
- [Can I persist data in a notebook?](/docs/phoenix/resources/frequently-asked-questions/can-i-persist-data-in-a-notebook.md): You can persist data in the notebook by either setting the usetempdir flag to false in px.launchapp which will persist your data in SQLite on your disk at the PHOENIX WORKING DIR. Alternatively you can deploy a phoenix instance and point to it via PHOENIX COLLECTOR ENDPOINT.
- [What is the difference between GRPC and HTTP?](/docs/phoenix/resources/frequently-asked-questions/what-is-the-difference-between-grpc-and-http.md): gRPC and HTTP are communication protocols used to transfer data between client and server applications.
- [Can I use gRPC for trace collection?](/docs/phoenix/resources/frequently-asked-questions/can-i-use-grpc-for-trace-collection.md): Phoenix does natively support gRPC for trace collection post 4.0 release. See Configuration for details.
- [How do I resolve Phoenix Evals showing NOT PARSABLE?](/docs/phoenix/resources/frequently-asked-questions/how-do-i-resolve-phoenix-evals-showing-not_parsable.md): NOTPARSABLE errors often occur when LLM responses exceed the maxtokens limit or produce incomplete JSON.
- [Braintrust Open Source Alternative? LLM Evaluation Platform Comparison](/docs/phoenix/resources/frequently-asked-questions/braintrust-open-source-alternative-llm-evaluation-platform-comparison.md): Braintrust is an evaluation platform that serves as an alternative to Arize Phoenix. Both platforms support core AI application needs, such as: evaluating AI applications, prompt management, tracing executions, and experimentation. However, there are a few major differences.
- [Langfuse alternative? Arize Phoenix vs Langfuse: Key differences](/docs/phoenix/resources/frequently-asked-questions/langfuse-alternative-arize-phoenix-vs-langfuse-key-differences.md): What is the difference between Arize Phoenix and Langfuse? Langfuse has an initially similar feature set to Arize Phoenix.
- [Open Source LangSmith Alternative: Arize Phoenix vs. LangSmith](/docs/phoenix/resources/frequently-asked-questions/open-source-langsmith-alternative-arize-phoenix-vs.-langsmith.md): What is the difference between Arize Phoenix and LangSmith?
- [Contribute to Phoenix](/docs/phoenix/resources/contribute-to-phoenix.md): If you want to contribute to the cutting edge of LLM and ML Observability, you've come to the right place!
- [Arize Phoenix or Arize AX](/docs/phoenix/resources/frequently-asked-questions/what-is-the-difference-between-phoenix-and-arize.md): How the open-source Phoenix platform and the commercial Arize AX platform differ, and how to choose between them.
- [Migrating to Arize AX](/docs/phoenix/resources/phoenix-to-arize-ax-migration.md): Migrate Phoenix data to Arize AX with the supported migration skill

## Integrations

- [Integrations](/docs/phoenix/integrations.md): Connect Phoenix with your favorite AI frameworks, LLM providers, and tools
- [Overview](/docs/phoenix/integrations/llm-providers/amazon-bedrock.md): Amazon Bedrock is a managed service that provides access to top AI models for building scalable applications.
- [Amazon Bedrock Tracing](/docs/phoenix/integrations/llm-providers/amazon-bedrock/amazon-bedrock-tracing.md): Instrument LLM calls to AWS Bedrock via the boto3 client using the BedrockInstrumentor
- [Amazon Bedrock SDK for JavaScript](/docs/phoenix/integrations/llm-providers/amazon-bedrock/amazon-bedrock-sdk-js.md): Instrument and observe AWS Bedrock calls using the AWS SDK for JavaScript
- [Amazon Bedrock Agents Tracing](/docs/phoenix/integrations/llm-providers/amazon-bedrock/amazon-bedrock-agents-tracing.md): Instrument LLM calls to AWS Bedrock via the boto3 client using the BedrockInstrumentor
- [Amazon Bedrock Agent Runtime JavaScript](/docs/phoenix/integrations/llm-providers/amazon-bedrock/amazon-bedrock-agent-runtime-js.md): Instrument and observe AWS Bedrock Agent Runtime calls in JavaScript/Node.js
- [Overview](/docs/phoenix/integrations/llm-providers/anthropic.md): Anthropic is an AI research company that develops LLMs, including Claude, with a focus on alignment and reliable behavior.
- [Anthropic Tracing](/docs/phoenix/integrations/llm-providers/anthropic/anthropic-tracing.md): Anthropic is a leading provider for state-of-the-art LLMs.
- [Anthropic SDK TypeScript](/docs/phoenix/integrations/llm-providers/anthropic/anthropic-sdk-typescript.md): Instrument and observe Anthropic calls in TypeScript/Node.js
- [Anthropic SDK Go](/docs/phoenix/integrations/llm-providers/anthropic/anthropic-sdk-go.md): Instrument and observe Anthropic calls in Go
- [Overview](/docs/phoenix/integrations/llm-providers/cohere.md): Cohere builds enterprise-grade large language models for chat, search, and retrieval-augmented generation.
- [Cohere Tracing](/docs/phoenix/integrations/llm-providers/cohere/cohere-tracing.md): Instrument LLM calls made using Cohere's Python client via the CohereInstrumentor
- [Overview](/docs/phoenix/integrations/llm-providers/google-gen-ai.md): Google GenAI is a suite of AI tools and models from Google Cloud, designed to help businesses build, deploy, and scale AI applications.
- [Google Gen AI Tracing](/docs/phoenix/integrations/llm-providers/google-gen-ai/google-genai-tracing.md): Instrument LLM calls made using the Google Gen AI Python SDK
- [Gemini Go SDK](/docs/phoenix/integrations/llm-providers/google-gen-ai/gemini-go-sdk.md): Instrument Gemini calls in Go using OpenTelemetry GenAI conventions
- [Overview](/docs/phoenix/integrations/llm-providers/groq.md): Groq provides ultra-low latency inference for LLMs through its custom-built LPU™ architecture.
- [Groq Tracing](/docs/phoenix/integrations/llm-providers/groq/groq-tracing.md): Instrument LLM applications built with Groq
- [Overview](/docs/phoenix/integrations/llm-providers/litellm.md): LiteLLM is an open-source platform that provides a unified interface to manage and access over 100 LLMs from various providers.
- [LiteLLM Tracing](/docs/phoenix/integrations/llm-providers/litellm/litellm-tracing.md): LiteLLM allows developers to call all LLM APIs using the openAI format. LiteLLM Proxy is a proxy server to call 100+ LLMs in OpenAI format.
- [Overview](/docs/phoenix/integrations/llm-providers/mistralai.md): Mistral AI develops open-weight large language models, focusing on efficiency, customization, and cost-effective AI solutions.
- [MistralAI Tracing](/docs/phoenix/integrations/llm-providers/mistralai/mistralai-tracing.md): Instrument LLM calls made using MistralAI's SDK via the MistralAIInstrumentor
- [Overview](/docs/phoenix/integrations/llm-providers/ollama.md): Ollama lets you run open-weight large language models such as Llama, Gemma, and Qwen locally with a simple API.
- [Ollama Tracing](/docs/phoenix/integrations/llm-providers/ollama/ollama-tracing.md): Instrument LLM calls made with the Ollama Python client via the OllamaInstrumentor
- [Overview](/docs/phoenix/integrations/llm-providers/openai.md): OpenAI provides state-of-the-art LLMs for natural language understanding and generation.
- [OpenAI Tracing](/docs/phoenix/integrations/llm-providers/openai/openai-tracing.md): Phoenix provides auto-instrumentation for the OpenAI Python Library.
- [OpenAI Agents SDK Tracing](/docs/phoenix/integrations/llm-providers/openai/openai-agents-sdk-tracing.md): Use Phoenix and OpenAI Agents SDK for powerful multi-agent tracing
- [OpenAI Node.js SDK](/docs/phoenix/integrations/llm-providers/openai/openai-node-js-sdk.md): Instrument and observe OpenAI calls
- [OpenAI Go SDK](/docs/phoenix/integrations/llm-providers/openai/openai-go-sdk.md): Instrument and observe OpenAI calls in Go
- [Overview](/docs/phoenix/integrations/llm-providers/openrouter.md): OpenRouter is a platform that connects developers to multiple AI models through a unified API, making it easier to compare, switch between, and integrate different models.
- [OpenRouter Tracing](/docs/phoenix/integrations/llm-providers/openrouter/openai-tracing.md): Phoenix provides auto-instrumentation for OpenRouter through the OpenAI Python Library since OpenRouter provides a fully OpenAI-compatible API endpoint.
- [Overview](/docs/phoenix/integrations/llm-providers/orcarouter.md): OrcaRouter is an OpenAI-compatible AI gateway that routes requests across 200+ models. Use Phoenix tracing to observe which upstream model served each request.
- [OrcaRouter Tracing](/docs/phoenix/integrations/llm-providers/orcarouter/openai-tracing.md): Phoenix provides auto-instrumentation for OrcaRouter through the OpenAI Python Library since OrcaRouter provides a fully OpenAI-compatible API endpoint.
- [Overview](/docs/phoenix/integrations/llm-providers/cloudflare-ai-gateway.md): Send traces from Cloudflare AI Gateway to Phoenix by exporting the gateway's OpenTelemetry spans or by instrumenting your app with OpenInference.
- [Tracing](/docs/phoenix/integrations/llm-providers/cloudflare-ai-gateway/cloudflare-ai-gateway-tracing.md): Configure Cloudflare AI Gateway to send traces to Phoenix by exporting the gateway's OpenTelemetry spans or by instrumenting your app with OpenInference.
- [Overview](/docs/phoenix/integrations/llm-providers/together.md): Together AI provides fast, cost-effective inference for a wide range of open-source models through a unified API.
- [Together AI Tracing](/docs/phoenix/integrations/llm-providers/together/together-tracing.md): Instrument LLM calls made using the Together AI SDK via the TogetherInstrumentor
- [Overview](/docs/phoenix/integrations/llm-providers/typesafe.md): TypeSafe AI is a decision-model provider that answers typed questions about structured state, returning schema-validated results instead of free-form text. Phoenix traces its calls as DECISION spans.
- [TypeSafe AI Tracing (Python)](/docs/phoenix/integrations/llm-providers/typesafe/typesafe-python.md): Instrument TypeSafe AI SDK calls in Python
- [TypeSafe AI Tracing (TypeScript)](/docs/phoenix/integrations/llm-providers/typesafe/typesafe-typescript.md): Instrument TypeSafe AI SDK calls in TypeScript/Node.js
- [Overview](/docs/phoenix/integrations/llm-providers/vertexai.md): Vertex AI is a fully managed platform by Google Cloud for building, deploying, and scaling machine learning models.
- [VertexAI Tracing](/docs/phoenix/integrations/llm-providers/vertexai/vertexai-tracing.md): Instrument LLM calls made using VertexAI's SDK via the VertexAIInstrumentor
- [Overview](/docs/phoenix/integrations/typescript/beeai.md): BeeAI is an open-source platform that enables developers to discover, run, and compose AI agents from any framework, facilitating the creation of interoperable multi-agent systems
- [BeeAI Tracing (JS)](/docs/phoenix/integrations/typescript/beeai/beeai-tracing-js.md): Auto-instrument and observe BeeAI agents
- [Claude Agent SDK](/docs/phoenix/integrations/typescript/claude-agent-sdk.md): Trace Anthropic's Claude Agent SDK applications in TypeScript/Node.js with Phoenix
- [Overview](/docs/phoenix/integrations/typescript/langchain.md): LangChain is an open-source framework for building language model applications with prompt chaining, memory, and external integrations
- [Langchain.js](/docs/phoenix/integrations/typescript/langchain/langchain-js.md): ![](https://www.npmjs.com/package/@arizeai/openinference-instrumentation-langchain) This module provides automatic instrumentation for LangChain.js,…
- [Overview](/docs/phoenix/integrations/typescript/mastra.md): Mastra is an open-source TypeScript AI agent framework designed for building production-ready AI applications with agents, workflows, RAG, and observability
- [Mastra Tracing](/docs/phoenix/integrations/typescript/mastra/mastra-tracing.md): Instrument agent applications built with Mastra
- [Overview](/docs/phoenix/integrations/typescript/mcp.md): Model Context Protocol (MCP) is an open protocol that enables secure connections between AI assistants and data sources.
- [MCP Tracing (TypeScript)](/docs/phoenix/integrations/typescript/mcp/mcp-tracing-typescript.md): Trace MCP (Model Context Protocol) clients and servers in TypeScript/Node.js
- [OpenAI Agents SDK](/docs/phoenix/integrations/typescript/openai-agents.md): Trace OpenAI Agents SDK applications in TypeScript/Node.js with Phoenix
- [Overview](/docs/phoenix/integrations/typescript/tanstack-ai.md): TanStack AI is a TypeScript framework for building chat, tool-calling, and agent-loop experiences across any model provider with a unified streaming API
- [TanStack AI Tracing](/docs/phoenix/integrations/typescript/tanstack-ai/tanstack-ai-tracing.md): Instrument TanStack AI chat, tool-calling, and agent loops with OpenInference middleware
- [Overview](/docs/phoenix/integrations/typescript/vercel.md): Vercel is a cloud platform that simplifies building, deploying, and scaling modern web applications with features like serverless functions, edge caching, and seamless Git integration
- [Vercel AI SDK Tracing (JS)](/docs/phoenix/integrations/typescript/vercel/vercel-ai-sdk-tracing-js.md): Trace Vercel AI SDK calls with @arizeai/phoenix-otel and send spans to Phoenix for LLM and agent observability.
- [Vercel Eve](/docs/phoenix/integrations/typescript/vercel/eve-tracing.md): Trace Vercel Eve agents with OpenInference and send AI SDK spans to Phoenix for LLM and agent observability.
- [Overview](/docs/phoenix/integrations/python/ag2.md): AG2 is an open-source Python framework for building multi-agent LLM applications, with agents, tools, middleware, and multi-agent networks
- [AG2 Tracing](/docs/phoenix/integrations/python/ag2/ag2-tracing.md): Auto-instrument your AG2 multi-agent application for seamless observability
- [Overview](/docs/phoenix/integrations/python/agentspec.md): Open Agent Spec (Agent Spec) is a portable language for defining agentic systems. It defines building blocks for standalone agents and structured agentic workflows as well as common ways of composing them into multi-agent systems.
- [Agent Spec Tracing](/docs/phoenix/integrations/python/agentspec/agentspec-tracing.md): How to use the python AgentSpecInstrumentor to trace Agent Spec workflows
- [Overview](/docs/phoenix/integrations/python/agno.md): Agno is an open-source Python framework for building lightweight, model-agnostic AI agents with built-in memory, knowledge, tools, and reasoning capabilities
- [Agno Tracing](/docs/phoenix/integrations/python/agno/agno-tracing.md): Phoenix provides seamless observability and tracing for Agno agents through the OpenInference instrumentation package. This integration automatically captures agent interactions, tool usage, reasoning steps, and multi-agent conversations, giving you complete visibility into your Agno applications. Monitor performance, debug issues, and evaluate agent behavior in real-time as your agents execute complex workflows and collaborate in teams.
- [Overview](/docs/phoenix/integrations/python/atomic-agents.md): Atomic Agents is a lightweight, modular Python framework for building agentic AI applications from small, schema-driven components on top of Instructor and Pydantic
- [Atomic Agents Tracing](/docs/phoenix/integrations/python/atomic-agents/atomic-agents-tracing.md): Trace Atomic Agents agents and tools in Phoenix: an AGENT span per agent run and a TOOL span per tool call, with the LLM calls nested underneath.
- [Overview](/docs/phoenix/integrations/python/autogen.md): AutoGen is an open-source Python framework for orchestrating multi-agent LLM interactions with shared memory and tool integrations to build scalable AI workflows
- [AutoGen Tracing](/docs/phoenix/integrations/python/autogen/autogen-tracing.md): openinference-instrumentation-autogen is now a thin, deprecated compatibility facade that delegates to AG2Instrumentor.
- [AutoGen AgentChat Tracing](/docs/phoenix/integrations/python/autogen/autogen-agentchat-tracing.md): Auto-instrument your AgentChat application for seamless observability
- [Overview](/docs/phoenix/integrations/python/beeai.md): BeeAI is an open-source platform that enables developers to discover, run, and compose AI agents from any framework, facilitating the creation of interoperable multi-agent systems
- [BeeAI Tracing (Python)](/docs/phoenix/integrations/python/beeai/beeai-tracing-python.md): Instrument and observe BeeAI agents
- [Claude Agent SDK](/docs/phoenix/integrations/python/claude-agent-sdk.md): Trace Anthropic's Claude Agent SDK applications in Python with Phoenix
- [Overview](/docs/phoenix/integrations/python/crewai.md): CrewAI is an open-source Python framework for orchestrating role-playing, autonomous AI agents into collaborative "crews" and "flows," combining high-level simplicity with fine-grained control.
- [CrewAI Tracing](/docs/phoenix/integrations/python/crewai/crewai-tracing.md): Instrument multi-agent applications using CrewAI
- [Overview](/docs/phoenix/integrations/python/dspy.md): DSPy is an open-source Python framework for declaratively programming modular LLM pipelines and automatically optimizing prompts and model weights
- [DSPy Tracing](/docs/phoenix/integrations/python/dspy/dspy-tracing.md): Instrument and observe your DSPy application via the DSPyInstrumentor
- [Overview](/docs/phoenix/integrations/python/google-adk.md): Google ADK is a Python SDK for building AI applications with Google's Gemini models and agent framework capabilities
- [Google ADK Tracing](/docs/phoenix/integrations/python/google-adk/google-adk-tracing.md): Instrument LLM calls made using the Google ADK Python SDK
- [Overview](/docs/phoenix/integrations/python/graphite.md): Graphite is an open-source Python framework for designing and orchestrating multi-agent LLM workflows with a visual builder and node-based composition.
- [Graphite Integration Guide](/docs/phoenix/integrations/python/graphite/graphite-integration-guide.md): This document provides comprehensive guidance on integrating Graphite with Phoenix for distributed tracing and observability.
- [Overview](/docs/phoenix/integrations/python/guardrails-ai.md): Guardrails is an open-source Python framework for adding programmable input/output validators to LLM applications, ensuring safe, structured, and compliant model interactions
- [Guardrails AI Tracing](/docs/phoenix/integrations/python/guardrails-ai/guardrails-ai-tracing.md): Instrument LLM applications that use the Guardrails AI framework
- [Overview](/docs/phoenix/integrations/python/haystack.md): Haystack is an open-source framework for building scalable semantic search and QA pipelines with document indexing, retrieval, and reader components
- [Haystack Tracing](/docs/phoenix/integrations/python/haystack/haystack-tracing.md): Instrument LLM applications built with Haystack
- [Overview](/docs/phoenix/integrations/python/hugging-face-smolagents.md): Hugging Face smolagents is a minimalist Python library for building powerful AI agents with simple abstractions, tool integrations, and flexible LLM support
- [Smolagents Tracing](/docs/phoenix/integrations/python/hugging-face-smolagents/smolagents-tracing.md): How to use the SmolagentsInstrumentor to trace smolagents by Hugging Face
- [Overview](/docs/phoenix/integrations/python/instructor.md): Instructor is a library that helps you define structured output formats for LLMs.
- [Instructor Tracing](/docs/phoenix/integrations/python/instructor/instructor-tracing.md): Install Be sure you also install the OpenInference library for the underlying model you're using along with Instructor.
- [Overview](/docs/phoenix/integrations/python/llamaindex.md): LlamaIndex is an open-source framework that streamlines connecting, ingesting, indexing, and retrieving structured or unstructured data to power efficient, data-aware language model applications.
- [LlamaIndex Tracing](/docs/phoenix/integrations/python/llamaindex/llamaindex-tracing.md): How to use the python LlamaIndexInstrumentor to trace LlamaIndex
- [LlamaIndex Workflows Tracing](/docs/phoenix/integrations/python/llamaindex/llamaindex-workflows-tracing.md): How to use the python LlamaIndexInstrumentor to trace LlamaIndex Workflows
- [Overview](/docs/phoenix/integrations/python/langchain.md): LangChain is an open-source framework for building language model applications with prompt chaining, memory, and external integrations
- [LangChain Tracing](/docs/phoenix/integrations/python/langchain/langchain-tracing.md): How to use the python LangChainInstrumentor to trace LangChain
- [Overview](/docs/phoenix/integrations/python/langgraph.md): LangGraph is an open-source framework for building graph-based LLM pipelines with modular nodes and seamless data integrations
- [LangGraph Tracing](/docs/phoenix/integrations/python/langgraph/langgraph-tracing.md): Phoenix has first-class support for LangGraph applications. LangGraph is supported by our LangChain instrumentor.
- [MCP Tracing](/docs/phoenix/integrations/python/mcp-tracing.md): Phoenix provides tracing for MCP clients and servers through OpenInference. This includes the unique capability to trace client to server interactions under a single trace in the correct hierarchy.
- [Overview](/docs/phoenix/integrations/python/nvidia.md): NVIDIA NeMo Agent Toolkit is a flexible, lightweight library for connecting enterprise agents to data sources and tools across any framework
- [NeMo Agent Toolkit Tracing](/docs/phoenix/integrations/python/nvidia/nemo-agent-tracing.md): Instrument and observe NVIDIA NeMo Agent Toolkit workflows
- [Overview](/docs/phoenix/integrations/python/portkey.md): Portkey is an AI Gateway and observability platform that provides routing, guardrails, caching, and monitoring for 200+ LLMs with enterprise-grade security and reliability features.
- [Portkey Tracing](/docs/phoenix/integrations/python/portkey/portkey-tracing.md): How to trace Portkey AI Gateway requests with Phoenix for comprehensive LLM observability
- [Overview](/docs/phoenix/integrations/python/pydantic.md): PydanticAI is a Python agent framework designed to make it less painful to build production-grade applications with Generative AI, built by the team behind Pydantic with type-safe structured outputs
- [Pydantic AI Tracing](/docs/phoenix/integrations/python/pydantic/pydantic-tracing.md): How to use the python PydanticAIInstrumentor to trace PydanticAI agents
- [Pydantic Evals](/docs/phoenix/integrations/python/pydantic/pydantic-evals.md): How to use Pydantic Evals with Phoenix to evaluate AI applications using structured evaluation frameworks
- [Overview](/docs/phoenix/integrations/python/restate.md): Restate is a durable execution platform that makes AI agents and workflows resumable and resilient, with built-in OpenTelemetry tracing
- [Restate Tracing](/docs/phoenix/integrations/python/restate/restate-tracing.md): Trace durable AI agent executions powered by Restate with Phoenix for full observability into LLM calls, tool invocations, and workflow steps.
- [Overview](/docs/phoenix/integrations/python/strands-agents.md): Strands Agents is an open-source AI agent SDK that uses model-driven orchestration to build production-ready agents in a few lines of code
- [Strands Agents Tracing](/docs/phoenix/integrations/python/strands-agents/strands-agents-tracing.md): Phoenix provides tracing support for Strands Agents through a span processor that transforms Strands' native OpenTelemetry spans into OpenInference format.
- [Overview](/docs/phoenix/integrations/java/google-adk.md): Trace Google Agent Development Kit applications written in Java with Phoenix.
- [Google ADK tracing for Java](/docs/phoenix/integrations/java/google-adk/google-adk-tracing.md): Instrument Google ADK for Java with OpenInference and export traces to Phoenix.
- [Overview](/docs/phoenix/integrations/java/langchain4j.md): LangChain4j is a Java library that provides APIs, tools, and patterns to easily build and integrate LLM-powered Java applications.
- [LangChain4j Tracing](/docs/phoenix/integrations/java/langchain4j/langchain4j-tracing.md): How to use OpenInference instrumentation with LangChain4j and export traces to Arize Phoenix.
- [Overview](/docs/phoenix/integrations/java/springai.md): Spring AI extends the Spring Framework, making it easier to integrate AI capabilities into Java applications using familiar Spring patterns.
- [Spring AI Tracing](/docs/phoenix/integrations/java/springai/springai-tracing.md): How to use OpenInference instrumentation with Spring AI and export traces to Arize Phoenix.
- [Overview](/docs/phoenix/integrations/java/arconia.md): Arconia is an open-source framework that you can add to an existing Spring Boot application to boost developer experience, reduce boilerplate, and seamlessly adopt cloud native patterns.
- [Arconia Tracing](/docs/phoenix/integrations/java/arconia/arconia-tracing.md): How to use OpenInference instrumentation for Spring AI with Arconia and export traces to Arize Phoenix.
- [Overview](/docs/phoenix/integrations/platforms/claude-managed-agents.md): Trace Anthropic Claude Managed Agents sessions in Phoenix by sending Claude Platform OpenTelemetry spans to a public Phoenix endpoint.
- [Claude Managed Agents](/docs/phoenix/integrations/platforms/claude-managed-agents/claude-managed-agents-tracing.md): Send OpenTelemetry traces of Claude Managed Agents sessions to Phoenix from the Claude Console and the Claude SDK, with turns, model requests, and tools.
- [Overview](/docs/phoenix/integrations/platforms/dify.md): Dify lets you visually build, orchestrate, and deploy AI-native apps using LLMs, with low-code workflows and agent frameworks for fast deployment.
- [Dify Tracing](/docs/phoenix/integrations/platforms/dify/dify-tracing.md): Configure your Dify application to view traces in Phoenix
- [Overview](/docs/phoenix/integrations/platforms/flowise.md): Flowise is a low-code platform for building customized chatflows and agentflows.
- [Flowise Tracing](/docs/phoenix/integrations/platforms/flowise/flowise-tracing.md): Analyzing and troubleshooting what happens under the hood can be challenging without proper insights. By integrating your Flowise application with Phoenix, you can monitor traces and gain robust observability into your chatflows and agentflows.
- [Overview](/docs/phoenix/integrations/platforms/langflow.md): Langflow is an open-source visual framework that enables developers to rapidly design, prototype, and deploy custom applications powered by large language models (LLMs)
- [LangFlow Tracing](/docs/phoenix/integrations/platforms/langflow/langflow-tracing.md): Pull Langflow Repo Navigate to the Langflow GitHub repo and pull the project down Create .env file Navigate to the repo and create a .env file with…
- [Overview](/docs/phoenix/integrations/platforms/prompt-flow.md): PromptFlow is a framework for designing, orchestrating, testing, and monitoring end-to-end LLM prompt workflows with built-in versioning and analytics
- [Prompt Flow Tracing](/docs/phoenix/integrations/platforms/prompt-flow/prompt-flow-tracing.md): Create flows using Microsoft PromptFlow and send their traces to Phoenix
- [Overview](/docs/phoenix/integrations/platforms/scalekit.md): Scalekit is an authentication and authorization platform for AI agents, handling delegated OAuth, token refresh, and scoped tool execution across its AgentKit connectors
- [Scalekit Tracing](/docs/phoenix/integrations/platforms/scalekit/scalekit-tracing.md): Trace Scalekit AgentKit tool calls in Phoenix.
- [Qdrant](/docs/phoenix/integrations/vector-databases/qdrant.md): Trace staged Qdrant hybrid search with Phoenix to isolate retrieval and selection issues across dense, sparse, and RRF fusion.
- [E2B](/docs/phoenix/integrations/sandboxes/e2b.md): Run Phoenix code evaluators in E2B's hosted micro-VM sandboxes.
- [Daytona](/docs/phoenix/integrations/sandboxes/daytona.md): Run Phoenix code evaluators in Daytona's managed development sandboxes.
- [Vercel Sandbox](/docs/phoenix/integrations/sandboxes/vercel.md): Run Phoenix code evaluators in ephemeral compute on Vercel's infrastructure.
- [Modal](/docs/phoenix/integrations/sandboxes/modal.md): Run Phoenix code evaluators in Modal's serverless container platform.
- [Docker Sandboxes](/docs/phoenix/integrations/sandboxes/docker.md): Run Phoenix code evaluators in Docker Cloud Sandboxes.
- [Cleanlab](/docs/phoenix/integrations/evaluation-integrations/cleanlab.md): Ensuring the reliability and accuracy of LLM-generated responses is a critical challenge for production AI systems.
- [MLflow](/docs/phoenix/integrations/evaluation-integrations/mlflow.md): Use Phoenix evaluators as MLflow scorers for GenAI evaluation workflows.
- [Ragas](/docs/phoenix/integrations/evaluation-integrations/ragas.md): This guide will walk you through the process of creating and evaluating agents using Ragas and Arize Phoenix.
- [UQLM Confidence & Hallucination Risk](/docs/phoenix/integrations/evaluation-integrations/uqlm.md): Ensuring reliable, accurate LLM responses is a core challenge in production AI.
- [Claude Code](/docs/phoenix/integrations/coding-agents/claude-code.md): Trace Claude Code CLI and Agent SDK sessions, tool usage, and token costs in Phoenix.
- [Codex](/docs/phoenix/integrations/coding-agents/codex.md): Trace OpenAI Codex CLI agent turns and tool calls in Phoenix.
- [GitHub Copilot](/docs/phoenix/integrations/coding-agents/copilot.md): Trace GitHub Copilot sessions in VS Code and the Copilot CLI in Phoenix.
- [Cursor](/docs/phoenix/integrations/coding-agents/cursor.md): Trace Cursor IDE and CLI conversations, shell commands, MCP tools, and file operations in Phoenix.
- [Gemini CLI](/docs/phoenix/integrations/coding-agents/gemini.md): Trace Gemini CLI sessions, agent steps, model calls, and tool invocations in Phoenix.
- [Kiro](/docs/phoenix/integrations/coding-agents/kiro.md): Trace Kiro CLI agent turns, tool calls, and credit usage in Phoenix.
- [Oh My Pi](/docs/phoenix/integrations/coding-agents/oh-my-pi.md): Trace Oh My Pi (omp) terminal coding sessions, model calls, tool usage, and token costs in Phoenix.
- [OpenCode](/docs/phoenix/integrations/coding-agents/opencode.md): Trace OpenCode terminal coding sessions, model calls, and tool usage in Phoenix.

## SDK & API Reference

- [Python SDK](/docs/phoenix/sdk-api-reference.md): Lightweight Python packages for tracing, evaluation, and platform interaction
- [arize-phoenix-client](/docs/phoenix/sdk-api-reference/python/arize-phoenix-client.md): Phoenix Client is a comprehensive SDK for interacting with the Phoenix
- [arize-phoenix-evals](/docs/phoenix/sdk-api-reference/python/arize-phoenix-evals.md): Tooling to evaluate LLM applications including faithfulness, correctness, relevance, and more.
- [arize-phoenix-otel](/docs/phoenix/sdk-api-reference/python/arize-phoenix-otel.md): Lightweight OpenTelemetry wrapper with Phoenix-aware defaults
- [TypeScript SDK](/docs/phoenix/sdk-api-reference/typescript/overview.md): Lightweight TypeScript packages for tracing, evaluation, and platform interaction
- [@arizeai/phoenix-client](/docs/phoenix/sdk-api-reference/typescript/arizeai-phoenix-client.md): ![NPM Version](https://www.npmjs.com/package/@arizeai/phoenix-client) TypeScript client for the Arize Phoenix API.
- [Overview](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/overview.md): Typed TypeScript client for Phoenix platform APIs
- [Projects](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/projects.md): List projects and manage project retention-policy and annotation-config assignments
- [Traces](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/traces.md): Retrieve, move, and annotate traces with @arizeai/phoenix-client
- [Spans](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/spans.md): Search and manage spans with @arizeai/phoenix-client
- [Annotations](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/annotations.md): Attach structured feedback to spans, documents, traces, and sessions with @arizeai/phoenix-client
- [Span Annotations](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/span-annotations.md): Log and retrieve span-level annotations with @arizeai/phoenix-client
- [Document Annotations](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/document-annotations.md): Log document-level annotations for RAG evaluation with @arizeai/phoenix-client
- [Session Annotations](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/session-annotations.md): Log session-level annotations for conversation evaluation with @arizeai/phoenix-client
- [Sessions](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/sessions.md): Work with sessions in @arizeai/phoenix-client
- [Secrets](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/secrets.md): Atomically manage encrypted provider credentials with the Phoenix TypeScript client
- [Users](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/users.md): Get the current Phoenix user with @arizeai/phoenix-client
- [Prompts](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/prompts.md): Manage prompts with @arizeai/phoenix-client
- [Datasets](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/datasets.md): Create and inspect datasets with @arizeai/phoenix-client
- [Experiments](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/experiments.md): Run experiments with @arizeai/phoenix-client
- [CI Eval Tests](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/ci-evals.md): Run dataset-backed Phoenix evaluations as Vitest or Jest tests
- [CI Eval Tests: Vitest](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/ci-evals-vitest.md): Wire @arizeai/phoenix-client/vitest into a Vitest project
- [CI Eval Tests: Jest](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/ci-evals-jest.md): Wire @arizeai/phoenix-client/jest into a Jest project
- [CI Eval Test Annotations](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-client/ci-evals-annotations.md): Record annotations and evaluator results on Phoenix experiment runs
- [@arizeai/phoenix-evals](/docs/phoenix/sdk-api-reference/typescript/arizeai-phoenix-evals.md): ![NPM Version](https://www.npmjs.com/package/@arizeai/phoenix-evals) TypeScript evaluation library for LLM applications.
- [Overview](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-evals/overview.md): Bundled docs for @arizeai/phoenix-evals
- [Create Evaluator](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-evals/create-evaluator.md): Build custom evaluators with @arizeai/phoenix-evals
- [LLM Evaluators](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-evals/llm-evaluators.md): Use LLM-backed evaluators in @arizeai/phoenix-evals
- [Classification Metrics](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-evals/classification-metrics.md): Precision, recall, and F-beta code evaluators in @arizeai/phoenix-evals
- [Templates](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-evals/templates.md): Template helpers in @arizeai/phoenix-evals
- [Classification](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-evals/classification.md): Classification helpers in @arizeai/phoenix-evals
- [Phoenix Integration](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-evals/phoenix-integration.md): Use @arizeai/phoenix-evals with Phoenix experiments
- [@arizeai/phoenix-otel](/docs/phoenix/sdk-api-reference/typescript/arizeai-phoenix-otel.md): Register OpenTelemetry with Phoenix and use the full OpenInference helper surface from one package
- [Overview](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-otel/overview.md): Register OpenTelemetry with Phoenix and use OpenInference helpers from @arizeai/phoenix-otel
- [Tracing Helpers](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-otel/tracing-helpers.md): Wrap functions and methods with OpenInference spans using withSpan, the span-kind wrappers (traceChain, traceAgent, traceTool, traceLLM, traceRetriever, traceReranker, traceEmbedding, traceGuardrail, traceEvaluator, tracePrompt), and observe
- [Context Attributes](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-otel/context-attributes.md): Propagate session IDs, user IDs, metadata, tags, prompt templates, and custom attributes to child spans
- [Manual Spans](/docs/phoenix/sdk-api-reference/typescript/packages/phoenix-otel/manual-spans.md): Create raw OpenTelemetry spans, build OpenInference attributes, and redact sensitive data with @arizeai/phoenix-otel
- [@arizeai/openinference-core](/docs/phoenix/sdk-api-reference/typescript/arizeai-openinference-core.md): OpenInference Core utilities for LLM tracing in TypeScript
- [@arizeai/phoenix-mcp](/docs/phoenix/sdk-api-reference/typescript/mcp-server.md): MCP server implementation for Arize Phoenix providing unified interface to Phoenix's capabilities.
- [@arizeai/phoenix-cli](/docs/phoenix/sdk-api-reference/typescript/arizeai-phoenix-cli.md): Command-line interface for retrieving trace data from Arize Phoenix projects.
- [REST API Overview](/docs/phoenix/sdk-api-reference/rest-api/overview.md): Start here to call the Phoenix REST API: endpoint setup, authentication, and your first request.
- [Create a session](/docs/phoenix/sdk-api-reference/rest-api/api-reference/agent-sessions/create-a-session.md)
- [List sessions](/docs/phoenix/sdk-api-reference/rest-api/api-reference/agent-sessions/list-sessions.md)
- [Get a session by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/agent-sessions/get-a-session-by-id.md)
- [Update a session](/docs/phoenix/sdk-api-reference/rest-api/api-reference/agent-sessions/update-a-session.md)
- [List session messages](/docs/phoenix/sdk-api-reference/rest-api/api-reference/agent-sessions/list-session-messages.md)
- [Submit tool outputs](/docs/phoenix/sdk-api-reference/rest-api/api-reference/agent-sessions/submit-tool-outputs.md)
- [Compact a session](/docs/phoenix/sdk-api-reference/rest-api/api-reference/agent-sessions/compact-a-session.md)
- [Chat with a session](/docs/phoenix/sdk-api-reference/rest-api/api-reference/agent-sessions/chat-with-a-session.md)
- [List annotation configurations](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotation-configs/list-annotation-configurations.md)
- [Create an annotation configuration](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotation-configs/create-an-annotation-configuration.md)
- [Get an annotation configuration by id or name](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotation-configs/get-an-annotation-configuration-by-id-or-name.md)
- [Update an annotation configuration](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotation-configs/update-an-annotation-configuration.md)
- [Delete an annotation configuration](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotation-configs/delete-an-annotation-configuration.md)
- [List annotation configurations assigned to a project](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotation-configs/list-annotation-configurations-assigned-to-a-project.md)
- [Assign an annotation configuration to a project](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotation-configs/assign-an-annotation-configuration-to-a-project.md)
- [Unassign an annotation configuration from a project](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotation-configs/unassign-an-annotation-configuration-from-a-project.md)
- [Replace the set of annotation configurations assigned to a project](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotation-configs/replace-the-set-of-annotation-configurations-assigned-to-a-project.md)
- [Get span annotations for a list of span ids](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotations/get-span-annotations-for-a-list-of-span_ids.md)
- [Get trace annotations for a list of trace ids](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotations/get-trace-annotations-for-a-list-of-trace_ids.md)
- [Get session annotations for a list of session ids](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotations/get-session-annotations-for-a-list-of-session_ids.md)
- [Delete span annotations by filter](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotations/delete-span-annotations-by-filter.md)
- [Delete trace annotations by filter](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotations/delete-trace-annotations-by-filter.md)
- [Delete session annotations by filter](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotations/delete-session-annotations-by-filter.md)
- [List the authenticated users api keys](/docs/phoenix/sdk-api-reference/rest-api/api-reference/api-keys/list-the-authenticated-users-api-keys.md)
- [Create an api key for the authenticated user](/docs/phoenix/sdk-api-reference/rest-api/api-reference/api-keys/create-an-api-key-for-the-authenticated-user.md)
- [Delete a user api key](/docs/phoenix/sdk-api-reference/rest-api/api-reference/api-keys/delete-a-user-api-key.md)
- [List all user api keys](/docs/phoenix/sdk-api-reference/rest-api/api-reference/api-keys/list-all-user-api-keys.md)
- [List system api keys](/docs/phoenix/sdk-api-reference/rest-api/api-reference/api-keys/list-system-api-keys.md)
- [Create a system api key](/docs/phoenix/sdk-api-reference/rest-api/api-reference/api-keys/create-a-system-api-key.md)
- [Delete a system api key](/docs/phoenix/sdk-api-reference/rest-api/api-reference/api-keys/delete-a-system-api-key.md)
- [Openai compatible chat completions](/docs/phoenix/sdk-api-reference/rest-api/api-reference/chat-completions/openai-compatible-chat-completions.md)
- [List datasets](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/list-datasets.md)
- [Get dataset by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/get-dataset-by-id.md)
- [Delete dataset by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/delete-dataset-by-id.md)
- [List dataset versions](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/list-dataset-versions.md)
- [Upload dataset from json csv or pyarrow](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/upload-dataset-from-json-csv-or-pyarrow.md)
- [Get examples from a dataset](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/get-examples-from-a-dataset.md)
- [Download dataset examples as csv file](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/download-dataset-examples-as-csv-file.md)
- [Download dataset examples as jsonl file](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/download-dataset-examples-as-jsonl-file.md)
- [Download dataset examples as openai fine tuning jsonl file](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/download-dataset-examples-as-openai-fine-tuning-jsonl-file.md)
- [Download dataset examples as openai evals jsonl file](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets/download-dataset-examples-as-openai-evals-jsonl-file.md)
- [List dataset labels](/docs/phoenix/sdk-api-reference/rest-api/api-reference/dataset-labels/list-dataset-labels.md)
- [Create a dataset label](/docs/phoenix/sdk-api-reference/rest-api/api-reference/dataset-labels/create-a-dataset-label.md)
- [Get a dataset label by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/dataset-labels/get-a-dataset-label-by-id.md)
- [Update a dataset label by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/dataset-labels/update-a-dataset-label-by-id.md)
- [Delete a dataset label by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/dataset-labels/delete-a-dataset-label-by-id.md)
- [List the labels applied to a dataset](/docs/phoenix/sdk-api-reference/rest-api/api-reference/dataset-labels/list-the-labels-applied-to-a-dataset.md)
- [Replace the set of labels applied to a dataset](/docs/phoenix/sdk-api-reference/rest-api/api-reference/dataset-labels/replace-the-set-of-labels-applied-to-a-dataset.md)
- [Apply a label to a dataset](/docs/phoenix/sdk-api-reference/rest-api/api-reference/dataset-labels/apply-a-label-to-a-dataset.md)
- [Remove a label from a dataset](/docs/phoenix/sdk-api-reference/rest-api/api-reference/dataset-labels/remove-a-label-from-a-dataset.md)
- [List experiments by dataset](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/list-experiments-by-dataset.md)
- [Create experiment on a dataset](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/create-experiment-on-a-dataset.md)
- [Get experiment by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/get-experiment-by-id.md)
- [Update an experiment by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/update-an-experiment-by-id.md)
- [Delete experiment by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/delete-experiment-by-id.md)
- [List the tags applied to an experiment](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/list-the-tags-applied-to-an-experiment.md)
- [Assign a tag to an experiment](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/assign-a-tag-to-an-experiment.md)
- [Remove a tag from an experiment](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/remove-a-tag-from-an-experiment.md)
- [Download experiment runs as a json file](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/download-experiment-runs-as-a-json-file.md)
- [Download experiment runs as a csv file](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/download-experiment-runs-as-a-csv-file.md)
- [List runs for an experiment](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/list-runs-for-an-experiment.md)
- [Create run for an experiment](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/create-run-for-an-experiment.md)
- [Get incomplete runs for an experiment](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/get-incomplete-runs-for-an-experiment.md)
- [Create or update evaluation for an experiment run](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/create-or-update-evaluation-for-an-experiment-run.md)
- [Get incomplete evaluations for an experiment](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments/get-incomplete-evaluations-for-an-experiment.md)
- [List traces for a project](/docs/phoenix/sdk-api-reference/rest-api/api-reference/traces/list-traces-for-a-project.md)
- [Delete traces from a project](/docs/phoenix/sdk-api-reference/rest-api/api-reference/traces/delete-traces-from-a-project.md)
- [Delete a trace by identifier](/docs/phoenix/sdk-api-reference/rest-api/api-reference/traces/delete-a-trace-by-identifier.md)
- [Create trace annotations](/docs/phoenix/sdk-api-reference/rest-api/api-reference/traces/create-trace-annotations.md)
- [Create a trace note](/docs/phoenix/sdk-api-reference/rest-api/api-reference/traces/create-a-trace-note.md)
- [Search spans with simple filters no dsl](/docs/phoenix/sdk-api-reference/rest-api/api-reference/spans/search-spans-with-simple-filters-no-dsl.md)
- [List spans with simple filters no dsl](/docs/phoenix/sdk-api-reference/rest-api/api-reference/spans/list-spans-with-simple-filters-no-dsl.md)
- [Create spans](/docs/phoenix/sdk-api-reference/rest-api/api-reference/spans/create-spans.md)
- [Create span annotations](/docs/phoenix/sdk-api-reference/rest-api/api-reference/spans/create-span-annotations.md)
- [Create a span note](/docs/phoenix/sdk-api-reference/rest-api/api-reference/spans/create-a-span-note.md)
- [Annotate span documents](/docs/phoenix/sdk-api-reference/rest-api/api-reference/spans/annotate-span-documents.md)
- [Delete a span by span identifier](/docs/phoenix/sdk-api-reference/rest-api/api-reference/spans/delete-a-span-by-span_identifier.md)
- [List all prompts](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/list-all-prompts.md)
- [Create a new prompt](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/create-a-new-prompt.md)
- [Delete a prompt](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/delete-a-prompt.md)
- [List prompt versions](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/list-prompt-versions.md)
- [Get prompt version by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/get-prompt-version-by-id.md)
- [Get prompt version by tag](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/get-prompt-version-by-tag.md)
- [Get latest prompt version](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/get-latest-prompt-version.md)
- [List prompt version tags](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/list-prompt-version-tags.md)
- [Add tag to prompt version](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/add-tag-to-prompt-version.md)
- [Delete a tag from a prompt version](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts/delete-a-tag-from-a-prompt-version.md)
- [List all projects](/docs/phoenix/sdk-api-reference/rest-api/api-reference/projects/list-all-projects.md)
- [Create a new project](/docs/phoenix/sdk-api-reference/rest-api/api-reference/projects/create-a-new-project.md)
- [Get project by id or name](/docs/phoenix/sdk-api-reference/rest-api/api-reference/projects/get-project-by-id-or-name.md)
- [Update a project by id or name](/docs/phoenix/sdk-api-reference/rest-api/api-reference/projects/update-a-project-by-id-or-name.md)
- [Delete a project by id or name](/docs/phoenix/sdk-api-reference/rest-api/api-reference/projects/delete-a-project-by-id-or-name.md)
- [List project sessions](/docs/phoenix/sdk-api-reference/rest-api/api-reference/sessions/list-project-sessions.md)
- [Get session by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/sessions/get-session-by-id.md)
- [Delete a session by identifier](/docs/phoenix/sdk-api-reference/rest-api/api-reference/sessions/delete-a-session-by-identifier.md)
- [Bulk delete sessions](/docs/phoenix/sdk-api-reference/rest-api/api-reference/sessions/bulk-delete-sessions.md)
- [Create session annotations](/docs/phoenix/sdk-api-reference/rest-api/api-reference/sessions/create-session-annotations.md)
- [List session annotations](/docs/phoenix/sdk-api-reference/rest-api/api-reference/sessions/list-session-annotations.md)
- [Create a session note](/docs/phoenix/sdk-api-reference/rest-api/api-reference/sessions/create-a-session-note.md)
- [Upsert or delete secrets](/docs/phoenix/sdk-api-reference/rest-api/api-reference/secrets/upsert-or-delete-secrets.md)
- [List all users](/docs/phoenix/sdk-api-reference/rest-api/api-reference/users/list-all-users.md)
- [Create a new user](/docs/phoenix/sdk-api-reference/rest-api/api-reference/users/create-a-new-user.md)
- [Delete a user by id](/docs/phoenix/sdk-api-reference/rest-api/api-reference/users/delete-a-user-by-id.md)
- [Get the authenticated user](/docs/phoenix/sdk-api-reference/rest-api/api-reference/users/get-the-authenticated-user.md)
- [OpenInference Java](/docs/phoenix/sdk-api-reference/openinference-sdk/openinference-java.md): Maven Central Repository: https://central.sonatype.com/search?q=arize The OpenInference Java SDK provides tracing capabilities for AI applications…

## Self-Hosting

- [Self-Hosting](/docs/phoenix/self-hosting.md): How to self-host a Phoenix instance
- [Architecture](/docs/phoenix/self-hosting/architecture.md): Phoenix is a containerized application that collects and analyzes traces from your LLM applications.
- [License](/docs/phoenix/self-hosting/license.md): Arize Phoenix is fully open-source free to self-host
- [Overview](/docs/phoenix/self-hosting/configuration.md): How to customize your self-hosted deployment of Phoenix
- [Using Amazon Aurora](/docs/phoenix/self-hosting/configuration/using-amazon-aurora.md): Phoenix supports IAM database authentication for PostgreSQL connections to Amazon Aurora/RDS.
- [Using Azure Database for PostgreSQL](/docs/phoenix/self-hosting/configuration/using-azure-managed-identity.md): Phoenix supports Microsoft Entra managed-identity authentication for Azure Database for PostgreSQL.
- [Terminal](/docs/phoenix/self-hosting/deployment-options/terminal.md): Run Phoenix via the CLI on your local machine
- [Docker](/docs/phoenix/self-hosting/deployment-options/docker.md): Deploy using docker compose for a local or cloud deployment
- [Kubernetes (kustomize)](/docs/phoenix/self-hosting/deployment-options/kubernetes.md): Phoenix can be deployed on Kubernetes with PostgreSQL using kustomize.
- [Kubernetes (helm)](/docs/phoenix/self-hosting/deployment-options/kubernetes-helm.md): Deploy Phoenix via Helm
- [AWS with CloudFormation](/docs/phoenix/self-hosting/deployment-options/aws-with-cloudformation.md): Phoenix can be deployed on AWS Fargate using CloudFormation
- [Railway](/docs/phoenix/self-hosting/deployment-options/railway.md): Use this guide to deploy Arize Phoenix on Railway via the prebuilt template.
- [Render](/docs/phoenix/self-hosting/deployment-options/render.md): Use this guide to deploy Arize Phoenix on Render via the prebuilt blueprint.
- [Google Cloud Run](/docs/phoenix/self-hosting/deployment-options/google-cloud-run.md): Use this guide to deploy Arize Phoenix on Google Cloud Run via the deploy button.
- [Azure](/docs/phoenix/self-hosting/deployment-options/azure.md): Use this guide to deploy Arize Phoenix on Azure Container Instances via the prebuilt ARM template.
- [Provisioning](/docs/phoenix/self-hosting/features/provisioning.md): How to provision Phoenix at deploy time
- [Authentication](/docs/phoenix/self-hosting/features/authentication.md): By default Phoenix deploys with authentication disabled as you may be just trying Phoenix for the very first time or have Phoenix deployed in a VPC.
- [Email](/docs/phoenix/self-hosting/features/email.md): Optionally, you can configure an SMTP server to send transactional emails. These are used for password resets, user invitations, and more.
- [Management](/docs/phoenix/self-hosting/features/management.md): How to manage your Phoenix instance
- [Sandbox Backends](/docs/phoenix/self-hosting/features/sandbox-runtimes.md): Set up the sandbox providers that run code evaluators on your self-hosted Phoenix.
- [Migrations](/docs/phoenix/self-hosting/upgrade/migrations.md): How Phoenix manages database migrations and how to run them safely in production
- [Privacy](/docs/phoenix/self-hosting/security/privacy.md): Understanding privacy and telemetry in self-hosted Phoenix deployments
- [Network Security](/docs/phoenix/self-hosting/security/network-security.md): Understanding outbound network requests and how to secure your self-hosted Phoenix deployment
- [Encryption](/docs/phoenix/self-hosting/advanced/encryption.md): Encryption in transit (TLS/HTTPS), at rest, and application-level encryption for self-hosted Phoenix deployments
- [FAQs](/docs/phoenix/self-hosting/misc/frequently-asked-questions.md): Permission denied writing to disc Some phoenix containers run as nonroot and therefore must be granted explicit write permissions to the mounted disc…

## Cookbooks

- [Cookbooks](/docs/phoenix/cookbook.md): Cookbooks & tutorials to help you build with Phoenix
- [Overview](/docs/phoenix/cookbook/agent-workflow-patterns.md): Workflows are the backbone of many successful LLM applications. They define how language models interact with tools, data, and users—often through a sequence of clearly orchestrated steps. Unlike fully autonomous agents, workflows offer structure and predictability, making them a practical choice for many real-world tasks.
- [AutoGen](/docs/phoenix/cookbook/agent-workflow-patterns/autogen.md): Use Phoenix to trace and evaluate AutoGen agents
- [CrewAI](/docs/phoenix/cookbook/agent-workflow-patterns/crewai.md): CrewAI is an open-source framework for building and orchestrating collaborative AI agents that act like a team of specialized virtual employees. Built on LangChain, it enables users to define roles, goals, and workflows for each agent, allowing them to work together autonomously on complex tasks with minimal setup.
- [Google GenAI SDK (manual orchestration)](/docs/phoenix/cookbook/agent-workflow-patterns/google-genai-sdk-manual-orchestration.md): Everything you need to know about Google's GenAI framework
- [OpenAI Agents](/docs/phoenix/cookbook/agent-workflow-patterns/openai-agents.md): OpenAI-Agents is a lightweight Python library for building agentic AI apps. It includes a few abstractions:
- [LangGraph](/docs/phoenix/cookbook/agent-workflow-patterns/langgraph.md): Use Phoenix to trace and evaluate agent frameworks built using Langgraph
- [Smolagents](/docs/phoenix/cookbook/agent-workflow-patterns/smolagents.md): SmolAgents is a lightweight Python library for composing tool-using, task-oriented agents. This guide outlines common agent workflows we've implemented—covering routing, evaluation loops, task orchestration, and parallel execution. For each pattern, we include an overview, a reference notebook, and guidance on how to evaluate agent quality.
- [Iterative Evaluation & Experimentation Workflow (Python)](/docs/phoenix/cookbook/ai-engineering-workflows/iterative-evaluation-and-experimentation-workflow-python.md): Phoenix Tracing, Evaluating, and Experimentation Walkthrough
- [Iterative Evaluation & Experimentation Workflow (TypeScript)](/docs/phoenix/cookbook/ai-engineering-workflows/iterative-evaluation-and-experimentation-workflow-typescript.md): Phoenix Tracing, Evaluating, and Experimentation Walkthrough
- [Aligning LLM Evals with Human Feedback (TypeScript)](/docs/phoenix/cookbook/human-in-the-loop-workflows-annotations/aligning-llm-evals-with-human-annotations-typescript.md): In this tutorial, we’ll run a Mastra agent and build a custom evaluator for it. The goal is to understand the workflow for creating evaluators that align with specific use cases.
- [Prompt Optimization Techniques](/docs/phoenix/cookbook/prompt-engineering/prompt-optimization.md): This tutorial will use Phoenix to compare the performance of different prompt optimization techniques.
- [Run Repetition Experiments](/docs/phoenix/cookbook/ai-engineering-workflows/analyzing-customer-review-evals-with-repetition-experiments.md): Large Language Models (LLMs) are probabilistic; the same prompt can yield different outputs across runs.
- [Identifying High-Signal Traces in Production](/docs/phoenix/cookbook/tracing/identify-high-signal-traces.md): When you have millions of traces, which ones are worth investigating? This guide shows you the techniques for surfacing anomalies, regressions, and failure clusters that deserve your attention.
- [Agentic RAG Tracing](/docs/phoenix/cookbook/tracing/agentic-rag-tracing.md): This tutorial demonstrates building an agentic RAG system using LlamaIndex's ReAct agent framework combined with vector and SQL query tools.
- [Trace and Evaluate Claude Managed Agents](/docs/phoenix/cookbook/tracing/claude-managed-agents-guide.md): Build an issue-triage agent on Claude Managed Agents, trace each session in Phoenix with OpenTelemetry, and evaluate the agent's triage decisions.
- [Generating Synthetic Datasets for LLM Evaluators & Agents](/docs/phoenix/cookbook/tracing/generating-synthetic-datasets-for-llm-evaluators-and-agents.md): Learn different strategies for dataset generation and show how they can be used to run experiments and test evaluators
- [Structured Data Extraction](/docs/phoenix/cookbook/tracing/structured-data-extraction.md): | Framework | Example notebook | | :---------------- | :--------------- | | Open AI Functions | | Overview Data extraction tasks using LLMs, such as…
- [Product Recommendation Agent: Google Agent Engine & LangGraph](/docs/phoenix/cookbook/tracing/product-recommendation-agent-google-agent-engine-and-langgraph.md): This notebook is adapted from Google's "Building and Deploying a LangGraph Application with Agent Engine in Vertex AI"
- [OpenInference Best Practices](/docs/phoenix/cookbook/tracing/openinference-best-practices.md): Learn OpenInference best practices for Phoenix and how to enrich auto-instrumented traces with LLM, tool, agent, chain, and session attributes.
- [More Cookbooks](/docs/phoenix/cookbook/tracing/cookbooks.md): 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.
- [Using Human Annotations for Eval Driven Development](/docs/phoenix/cookbook/human-in-the-loop-workflows-annotations/using-human-annotations-for-eval-driven-development.md): How to leverage human annotations to build evaluations and experiments that improve your system
- [Optimizing Coding Agent Prompts - Prompt Learning](/docs/phoenix/cookbook/prompt-engineering/optimizing-coding-agent-prompts-prompt-learning.md): Optimizing coding agent prompts and tracking coding agent improvement
- [Optimizing Prompts for LLM Classification - Prompt Learning](/docs/phoenix/cookbook/prompt-engineering/prompt-learning-optimizing-prompts-for-classification.md): Using Prompt Learning to boost accuracy on a classification dataset
- [Few Shot Prompting](/docs/phoenix/cookbook/prompt-engineering/few-shot-prompting.md): Few-shot prompting is a powerful technique in prompt engineering that helps LLMs perform tasks more effectively by providing a few examples within the prompt.
- [ReAct Prompting](/docs/phoenix/cookbook/prompt-engineering/react-prompting.md): ReAct (Reasoning + Acting) is a prompting technique that enables LLMs to think step-by-step before taking action. Unlike traditional prompting, where a model directly provides an answer, ReAct prompts guide the model to reason through a problem first, then decide which tools or actions are necessary to reach the best solution.
- [Chain of Thought Prompting](/docs/phoenix/cookbook/prompt-engineering/chain-of-thought-prompting.md): colab.research.google.com LLMs excel at text generation, but their reasoning abilities depend on how we prompt them.
- [LLM-as-a-Judge Prompt Optimization](/docs/phoenix/cookbook/prompt-engineering/llm-as-a-judge-prompt-optimization.md): LLM as a Judge An LLM as a Judge refers to using an LLM as a tool for evaluating and scoring responses based on predefined criteria.
- [Write Your First Custom Eval](/docs/phoenix/cookbook/evaluation/creating-a-custom-llm-evaluator-with-a-benchmark-dataset.md): Learn how to build a custom LLM-as-a-Judge evaluator by creating a benchmark dataset tailored to your use case, enabling rigorous evaluation beyond standard templates.
- [OpenAI Agents SDK Cookbook](/docs/phoenix/cookbook/evaluation/openai-agents-sdk-cookbook.md): This guide shows you how to create and evaluate agents with Phoenix to improve performance.
- [Evaluate a Talk-to-Your-Data Agent](/docs/phoenix/cookbook/evaluation/evaluate-an-agent.md): colab.research.google.com This notebook serves as an end-to-end example of how to trace and evaluate an agent.
- [Trace-level Evaluation: Beyond Input/Output Checks](/docs/phoenix/cookbook/evaluation/trace-level-evaluation.md): Evaluate an agent's intermediate reasoning, tool selection, and decision path — not just its final answer.
- [Session-level Evaluation: Scoring the Whole Conversation](/docs/phoenix/cookbook/evaluation/session-level-evaluation.md): Evaluate a multi-turn session as a whole - coherence, goal completion, and user frustration that turn-by-turn checks miss.
- [Evaluate RAG](/docs/phoenix/cookbook/evaluation/evaluate-rag.md): Building a RAG pipeline and evaluating it with Phoenix Evals.
- [Code Readability Evaluation](/docs/phoenix/cookbook/evaluation/code-readability-evaluation.md): Evaluate the readability of code generated by LLM applications using Phoenix's evaluation framework.
- [Jailbreak and Prompt Injection Defense](/docs/phoenix/cookbook/guardrails/jailbreak-and-prompt-injection-defense.md): Red-team a support assistant across a taxonomy of attacks, score Attack Success Rate, and find which defenses actually work, including the indirect injection that input filtering can't see.
- [Designing Realtime Guardrails: Input, Output, and the Cost of Blocking](/docs/phoenix/cookbook/guardrails/designing-realtime-guardrails.md): Decide what to guard at input vs. output, trade latency against coverage, and layer guardrails without blocking real users.
- [Analyzing Customer Review Evals with Repetition Experiments](/docs/phoenix/cookbook/datasets-and-experiments/analyzing-customer-review-evals-with-repetition-experiments.md): Large Language Models (LLMs) are probabilistic; the same prompt can yield different outputs across runs. This variability makes it hard to tell if a change truly improves performance or is just random noise.
- [Experiment with a Customer Support Agent](/docs/phoenix/cookbook/datasets-and-experiments/experiment-with-a-customer-support-agent.md): This guide shows you how to create and evaluate agents with Phoenix to improve performance.
- [Comparing LlamaIndex Query Engines with a Pairwise Evaluator](/docs/phoenix/cookbook/datasets-and-experiments/comparing-llamaindex-query-engines-with-a-pairwise-evaluator.md): This tutorial sets up an experiment to determine which LlamaIndex query engine is preferred by an evaluation LLM. Using the PairwiseEvaluator module, we compare responses from different engines and identify which one produces more helpful or relevant outputs.
- [Prompt Template Iteration for a Summarization Service](/docs/phoenix/cookbook/datasets-and-experiments/summarization.md): Imagine you're deploying a service for your media company's summarization model that condenses daily news into concise summaries to be displayed online. One challenge of using LLMs for summarization is that even the best models tend to be verbose.
- [Text2SQL Experiments](/docs/phoenix/cookbook/datasets-and-experiments/text2sql.md): Building effective text-to-SQL systems requires rigorous evaluation and systematic experimentation. In this tutorial, we'll walk through the complete evaluation-driven development process, starting from scratch without pre-existing datasets of questions or expected responses.
- [Why Public Benchmarks Lie: Building Your Own Eval Harness](/docs/phoenix/cookbook/datasets-and-experiments/building-your-own-eval-harness.md): A model that wins on MMLU can lose on your task. Build a domain-specific harness and compare models fairly on your own data and metric.
- [More Cookbooks](/docs/phoenix/cookbook/datasets-and-experiments/cookbooks.md): Iteratively improve your LLM task by building datasets, running experiments, and evaluating performance using code and LLM-as-a-Judge.

## Release Notes

- [Release Notes](/docs/phoenix/release-notes.md): The latest from the Phoenix team.
- [10.13.2026: Docker Sandboxes and MCP GraphQL Tools](/docs/phoenix/release-notes/10-2026/10-13-2026-docker-sandboxes-and-mcp-graphql-tools.md): Run code evaluators in Docker's hosted sandbox environment, test Claude Haiku 5.5 in the playground, and let MCP clients inspect Phoenix data through GraphQL.
- [10.01.2026: Trace Tree Evals and Decision Spans](/docs/phoenix/release-notes/10-2026/10-01-2026-trace-tree-evals-and-decision-spans.md): Find failed evals while scanning the trace tree, trace decision-model calls as DECISION spans, and manage project annotation configs from the TypeScript client.
- [09.29.2026: Editable Dataset Examples, Trace Metrics, and Custom Skills](/docs/phoenix/release-notes/09-2026/09-29-2026-editable-dataset-examples-and-trace-metrics.md): Edit dataset examples in the table and save the batch as one version, read latency, tokens, and cost on every trace tree row, and mount your own skills into MCP and PXI.
- [09.14.2026: Coding Agent Plugins and Error Analysis](/docs/phoenix/release-notes/09-2026/09-14-2026-coding-agent-plugins-and-error-analysis.md): Use Phoenix from your coding tools. Run error analysis through MCP and check whether a conversation covered every request.
- [09.08.2026: Harbor Plugin for Agent Benchmarks](/docs/phoenix/release-notes/09-2026/09-08-2026-harbor-plugin-for-agent-benchmarks.md): Record Harbor agent benchmark jobs in Phoenix as datasets and experiments, with verifier rewards as evaluations and ATIF trajectories as traces.
- [09.08.2026: New Model Providers, Trace Filters, and PXI Updates](/docs/phoenix/release-notes/09-2026/09-08-2026-model-providers-trace-filters-and-pxi.md): Run Z.ai, MiniMax, and Meta models in the playground and PXI, filter traces by error status and latency from both SDKs, and find PXI settings and slash commands faster.
- [09.01.2026: GitHub Issues from PXI, PII Detection, and Client Updates](/docs/phoenix/release-notes/09-2026/09-01-2026-github-issues-pii-detection-and-client-updates.md): PXI can file the GitHub issue for the bug it just found, a new evaluator screens conversations for PII, prompt versions land over REST, and the TypeScript client picks up prompt, trace, project, and user helpers.
- [08.25.2026: Smarter PXI Workflows and Retrieval Evaluation](/docs/phoenix/release-notes/08-2026/08-25-2026-pxi-browser-actions-and-retrieval-relevance.md): PXI completes multi-step UI workflows with approval-gated changes, a new evaluator scores retrievals from any source, and new trace and REST tools speed up analysis and administration.
- [08.17.2026: Trace Filter Expressions and Analytics SQL](/docs/phoenix/release-notes/08-2026/08-17-2026-trace-filters-and-analytics-sql.md): Filter the traces table with a full expression language, ask Phoenix arbitrary analytical questions with read-only SQL over MCP, read annotation details on hover, and filter spans by PXI approval decisions.
- [08.12.2026: Session Filter Expressions, REST API Expansion, and Endpoint Configuration](/docs/phoenix/release-notes/08-2026/08-12-2026-session-filters-rest-api-and-endpoint-config.md): Filter sessions with a full expression language and plain-English AI query, manage dataset splits and experiment tags over REST, move traces between projects, update prompt metadata from both SDKs, point every SDK at one PHOENIXENDPOINT, and log in to OAuth2 with a platform-minted assertion.
- [08.11.2026: Persistent Agent Sessions](/docs/phoenix/release-notes/08-2026/08-11-2026-persistent-agent-sessions.md): Conversations with the Phoenix agent are saved server-side — browse, restore, rename, rewind, and branch them from the browser or the terminal, interchangeably.
- [08.04.2026: Chat Completions Proxy, AI Query, and Annotation Charts](/docs/phoenix/release-notes/08-2026/08-04-2026-chat-completions-proxy-ai-query-and-annotation-charts.md): An OpenAI-compatible chat completions endpoint backed by server-held credentials, plain-English filter composition, per-annotation metric charts with no selection cap, a conversation-grounded Hallucination evaluator, annotations in span downloads, and a pinned note bar in span details.
- [07.28.2026: Experiment Charts, Span Downloads, and Root-Span Filters](/docs/phoenix/release-notes/07-2026/07-28-2026-experiment-charts-span-downloads-and-root-span-filters.md): Metric charts above the experiments table, span and trace downloads as OTLP JSON, span-ID and root-span filtering, span detail as searchable tables, a Toxicity evaluator, the Monty local sandbox, and Claude Opus 5.
- [07.22.2026: MCP Client Setup, Provider Filtering, and User Friction Evals](/docs/phoenix/release-notes/07-2026/07-22-2026-mcp-setup-provider-filter-and-evals.md): One-command MCP client setup and a Settings guide, a playground model picker scoped to provisioned providers, a new User Friction evaluator in both SDKs, and AI SDK v7 tracing.
- [07.17.2026: OAuth2 Authorization Server & Remote MCP](/docs/phoenix/release-notes/07-2026/07-17-2026-oauth2-authorization-server.md): Phoenix 19 turns Phoenix into its own OAuth2 authorization server: browser-based px CLI login, a built-in Remote MCP server, REST API-key management, and a tightened key-issuance policy.
- [07.14.2026: Command Palette, Session Tools, and Table Customization](/docs/phoenix/release-notes/07-2026/07-14-2026-command-palette-sessions-and-tables.md): Jump anywhere with the ⌘K command palette, customize and reorder table columns, inspect session stats, discover credentials from a .env.phoenix file, filter projects by name, and use the GPT 5.6 family in the Playground.
- [07.07.2026: Metric Charts, Trace Search, and REST API Expansion](/docs/phoenix/release-notes/07-2026/07-07-2026-metric-charts-trace-search-and-rest-api.md): Pin metric charts above your data tables, search the trace tree, manage dataset labels and annotation-config assignments over REST, run classification-metric evaluators in TypeScript, and use Claude Sonnet 5 in the Playground.
- [06.30.2026: PXI Terminal Client and Annotation Summary](/docs/phoenix/release-notes/06-2026/06-30-2026-pxi-terminal-client-and-annotation-summary.md): Run PXI as an interactive terminal chat from the Phoenix CLI, review and bulk-delete annotations from project settings, and copy trace IDs from experiment trace details.
- [06.26.2026: Evals as Tests — pytest and Vitest/Jest Integrations](/docs/phoenix/release-notes/06-2026/06-26-2026-eval-ci-pytest-vitest-jest.md): Write LLM evaluations as ordinary pytest, Vitest, or Jest tests that record every run to Phoenix and gate CI.
- [06.24.2026: Annotations, Labels, and PXI Server Tools](/docs/phoenix/release-notes/06-2026/06-24-2026-annotations-labels-and-pxi-server-tools.md): Trace-level annotations surfaced across the trace header and project stats, label management from the prompts and datasets lists, OAuth2 role-override preservation, and a server-side bash tool for PXI subagents.
- [06.16.2026: Time Range, Metrics, and Sharing Upgrades](/docs/phoenix/release-notes/06-2026/06-16-2026-time-range-metrics-and-sharing.md): A calendar time range picker with pan/zoom and live streaming controls, shareable trace URLs, trace and session annotation metrics, token detail charts, and opt-in PXI subagents.
- [06.11.2026: Time Range Selector and PXI Slash Commands](/docs/phoenix/release-notes/06-2026/06-11-2026-time-range-and-pxi-slash-commands.md): Search presets and type free-form durations in the time range selector; PXI adds local slash commands and dataset evaluator editing.
- [06.10.2026: PXI Agent Update](/docs/phoenix/release-notes/06-2026/06-10-2026-pxi-agent-update.md): PXI gains a skills menu, parallel subagents, full playground orchestration, evaluator authoring, and dataset management — plus Claude Fable 5 in the playground.
- [06.02.2026: Introducing PXI](/docs/phoenix/release-notes/06-2026/06-02-2026-pxi-agent.md): PXI (Phoenix Intelligence) is the AI engineering agent built into Phoenix — hand it an investigation instead of digging through traces, prompts, and experiments yourself.
- [05.27.2026: Drag-to-Zoom on Project Metric Charts](/docs/phoenix/release-notes/05-2026/05-27-2026-drag-to-zoom.md): Click and drag across any project metric chart or the spans sparkline to zoom into a selected time window.
- [05.21.2026: Code Evaluators](/docs/phoenix/release-notes/05-2026/05-21-2026-code-evaluators.md): Write Python or TypeScript evaluation functions directly in the Phoenix UI — no SDK, local runtime, or deploy step — and run them server-side on every experiment.
- [05.15.2026: OTel GenAI Semantic Convention Auto-Conversion](/docs/phoenix/release-notes/05-2026/05-15-2026-otel-semconv-conversion.md): Phoenix now automatically converts OpenTelemetry GenAI semantic convention attributes to OpenInference, so traces from OTel-native instrumentations render with full message I/O, tool calls, and token counts.
- [05.15.2026: Session Trace Feedback and ATIF v1.7 Support](/docs/phoenix/release-notes/05-2026/05-15-2026-session-feedback-and-atif.md): Trace Feedback in Session Turn View May 15, 2026 Available in arize-phoenix 15.10.0+ The Session Details view now includes a feedback toolbar on each…
- [05.13.2026: Playground Thinking Controls for Anthropic and Google](/docs/phoenix/release-notes/05-2026/05-13-2026-playground-thinking-controls.md): The Phoenix Playground now exposes first-class thinking controls for Anthropic extended thinking and Google thinking — token budgets, effort levels, and display toggles.
- [05.13.2026: Session Enhancements and Annotation Identifiers](/docs/phoenix/release-notes/05-2026/05-13-2026-session-and-annotations.md): Session turn messages are now expandable, notes accept caller-supplied identifiers for upsert semantics, and new CLI commands manage annotation bulk-delete.
- [05.10.2026: Playground Preferences](/docs/phoenix/release-notes/05-2026/05-10-2026-playground-preferences.md): Save a default provider and model for the Playground, and the metrics aside is now always visible on the project page.
- [05.08.2026: OTLP Project Routing via HTTP Header](/docs/phoenix/release-notes/05-2026/05-08-2026-otlp-project-header.md): Route OTLP traces to a named Phoenix project by setting the x-project-name HTTP header — no resource attribute required.
- [05.05.2026: Provider Tools in Playground and Prompts](/docs/phoenix/release-notes/05-2026/05-05-2026-provider-tools.md): Phoenix Playground and Prompts now support vendor-native tools — web search, code execution, computer use, grounding, and more — alongside existing function tools.
- [05.05.2026: REST API Updates](/docs/phoenix/release-notes/05-2026/05-05-2026-rest-api-updates.md): Filter-Based Annotation Delete Endpoints May 5, 2026 Available in arize-phoenix 15.4.0+ Three new DELETE endpoints let you bulk-remove annotations…
- [05.01.2026 TanStack AI Tracing](/docs/phoenix/release-notes/05-2026/05-01-2026-tanstack-ai-tracing.md): Instrument TanStack AI chat, tool-calling, and agent loops with the new @arizeai/openinference-tanstack-ai middleware.
- [04.30.2026: CLI Named Auth Profiles](/docs/phoenix/release-notes/04-2026/04-30-2026-cli-auth-profiles.md): px profile commands let you store named Phoenix connection profiles and switch between them without re-exporting environment variables.
- [04.30.2026: Annotation Enhancements](/docs/phoenix/release-notes/04-2026/04-30-2026-annotation-enhancements.md): Session Annotations and Notes (CLI and TypeScript) April 30, 2026 Available in arize-phoenix 15.1.0+ (server), @arizeai/phoenix-cli 1.4.0+ (CLI),…
- [04.29.2026: Dataset Upsert](/docs/phoenix/release-notes/04-2026/04-29-2026-dataset-upsert.md): createdataset now upserts — if a dataset with the same name exists, examples are updated in place rather than raising a conflict error.
- [04.28.2026 Session Notes API](/docs/phoenix/release-notes/04-2026/04-28-2026-session-notes-api.md): Add session notes through a dedicated REST endpoint and reserve the note annotation name for session note APIs.
- [04.24.2026 Trace Notes API](/docs/phoenix/release-notes/04-2026/04-24-2026-trace-notes-api.md): Add trace notes via REST endpoint, TypeScript client, or CLI; reserve the note annotation name for note-specific APIs.
- [04.24.2026 arize-phoenix-otel 0.16](/docs/phoenix/release-notes/04-2026/04-24-2026-phoenix-otel-python-0-16.md): OpenInference context managers and semantic conventions ship from a single phoenix.otel import — no second install required.
- [04.22.2026 Secrets Settings Page and Evaluator Trace ID](/docs/phoenix/release-notes/04-2026/04-22-2026-secrets-ui-and-evaluator-trace-id.md): Manage LLM provider secrets in the UI; pass trace IDs to experiment evaluators for correlation and debugging.
- [04.20.2026 Span Attribute Filtering, CLI Notes, and Claude Opus 4.7](/docs/phoenix/release-notes/04-2026/04-20-2026-span-attribute-filter-cli-notes-and-opus-4-7.md): Filter spans by attributes across Python, TypeScript, REST, and CLI; add notes to spans via px span add-note; use Claude Opus 4.7 in the Playground.
- [04.16.2026 Azure Managed Identity for PostgreSQL](/docs/phoenix/release-notes/04-2026/04-16-2026-azure-managed-identity-postgres.md): Connect Phoenix to Azure Database for PostgreSQL using managed identity — no static passwords required.
- [04.14.2026 CLI Annotation Commands](/docs/phoenix/release-notes/04-2026/04-14-2026-cli-annotation-commands.md): Write span and trace annotations from the terminal with px span annotate and px trace annotate.
- [04.13.2026 @arizeai/phoenix-otel 1.0](/docs/phoenix/release-notes/04-2026/04-13-2026-phoenix-otel-ts-1-0.md): Tracing helpers, decorators, context setters, and OpenInference semantic conventions ship from a single @arizeai/phoenix-otel import.
- [04.10.2026 Shareable Project URLs](/docs/phoenix/release-notes/04-2026/04-10-2026-shareable-url-redirects.md): Link to Phoenix projects by name without looking up internal IDs.
- [04.07.2026 PostgreSQL Read Replica Routing](/docs/phoenix/release-notes/04-2026/04-07-2026-postgresql-read-replica.md): Route read-only queries to a PostgreSQL read replica to reduce load on the primary under high ingestion.
- [Release Notes](/docs/phoenix/release-notes/04-2026/04-07-2026-phoenix-v14-breaking-changes.md): Breaking changes in Phoenix v14.0.0: CLI restructuring, legacy client removal, evaluations endpoint removal, evals 1.0 removal, and GraphQL pagination.
- [04.03.2026 ATIF Trajectory Upload](/docs/phoenix/release-notes/04-2026/04-03-2026-atif-trajectory-upload.md): Upload Harbor ATIF agent trajectories as structured Phoenix traces.
- [Release Notes](/docs/phoenix/release-notes/04-2026/04-01-2026-get-traces-secrets-api-and-python-314.md): gettraces — Retrieve Traces from a Project April 1, 2026 Available in arize-phoenix 13.15.0+ (server), arize-phoenix-client 2.2.0+ (Python)…
- [Release Notes](/docs/phoenix/release-notes/03-2026/03-30-2026-delete-prompts-api.md): Delete Prompts and Prompt Version Tags via REST API March 30, 2026 Available in arize-phoenix 13.20.0+ Two new REST endpoints let you delete prompts…
- [Release Notes](/docs/phoenix/release-notes/03-2026/03-24-2026-prompt-version-diff-and-evals-updates.md): Prompt Version Diff View March 24, 2026 Available in arize-phoenix 13.18.0+ The Prompts UI now includes a diff view for comparing two versions of a…
- [Release Notes](/docs/phoenix/release-notes/03-2026/03-22-2026-cli-and-user-api.md): px spans — Fetch and Filter Spans from the CLI March 22, 2026 Available in arize-phoenix 13.16.0+ and @arizeai/phoenix-cli 0.12.0+ The Phoenix CLI…
- [Release Notes](/docs/phoenix/release-notes/03-2026/03-13-2026-rest-api-improvements.md): List Traces by Project March 13, 2026 Available in arize-phoenix 13.15.0+ A new GET /v1/projects/{projectidentifier}/traces REST endpoint lists…
- [Release Notes](/docs/phoenix/release-notes/03-2026/03-11-2026-session-turns-api.md): Session Turns API March 11, 2026 Available in arize-phoenix-client 2.0.0+ (Python) and @arizeai/phoenix-client 6.4.0+ (TypeScript) Phoenix now…
- [03.08.2026 New Playground Providers and Project Settings](/docs/phoenix/release-notes/03-2026/03-08-2026-new-playground-providers-and-project-settings.md): Phoenix v13.10.0 adds Cerebras, Fireworks AI, Groq, and Moonshot as first-class playground providers, plus editable project settings.
- [03.05.2026 SDK Session Retrieval](/docs/phoenix/release-notes/03-2026/03-05-2026-sdk-session-retrieval.md): Get and list sessions programmatically from Python and TypeScript.
- [02.27.2026 Sessions API and CLI Support](/docs/phoenix/release-notes/02-2026/02-27-2026-cli-sessions-and-rest-api.md): Why Sessions Matter LLM applications are increasingly multi-turn.
- [Phoenix 13.0](/docs/phoenix/release-notes/02-2026/02-14-2026-phoenix-13-0.md): Dataset Evaluators, custom model providers, OpenAI Responses API support, and major Playground and experiment UX improvements.
- [01.17.2026 Phoenix CLI: Terminal Access for AI Coding Assistants](/docs/phoenix/release-notes/01-2026/01-17-2026-phoenix-cli-ai-agent-debugging.md): Full documentation Install from npm Overview @arizeai/phoenix-cli is a command-line interface for retrieving trace data from Phoenix.
- [12.20.2025: Improved User Preferences](/docs/phoenix/release-notes/12-2025/12-20-2025-improved-user-preferences.md): Available in Phoenix 12.27+
- [12.12.2025: Support for Gemini Tool Calls](/docs/phoenix/release-notes/12-2025/12-12-2025-support-for-gemini-tool-calls.md): Available in Phoenix 12.25+
- [12.09.2025: Span Notes API](/docs/phoenix/release-notes/12-2025/12-09-2025-span-notes-api.md): Available in Phoenix 12.21+
- [12.06.2025: LDAP Authentication Support](/docs/phoenix/release-notes/12-2025/12-06-2025-ldap-authentication-support.md): Available in Phoenix 12.20+
- [12.04.2025: Evaluator Message Formats](/docs/phoenix/release-notes/12-2025/12-04-2025-evaluator-message-formats.md): Available in phoenix-evals 0.22+ (Python) and @arizeai/phoenix-evals 2.0+ (TypeScript)
- [12.03.2025: TypeScript createEvaluator](/docs/phoenix/release-notes/12-2025/12-03-2025-typescript-create-evaluator.md): Available in @arizeai/phoenix-evals 2.0+
- [12.01.2025: Splits on Experiments Table](/docs/phoenix/release-notes/12-2025/12-01-2025-splits-on-experiments-table.md): Available in Phoenix 12.20+
- [11.29.2025: Add support for Claude Opus 4-5](/docs/phoenix/release-notes/11-2025/11-29-2025-add-support-for-claude-opus-4-5.md): Available in Phoenix 12.18+
- [11.27.2025: Show Server Credential Setup in Playground API Keys](/docs/phoenix/release-notes/11-2025/11-27-2025-show-server-credential-setup-in-playground-api-keys.md): Available in Phoenix 12.18+
- [11.25.2025: Split Assignments When Uploading a Dataset](/docs/phoenix/release-notes/11-2025/11-25-2025-split-assignments-when-uploading-a-dataset.md): Available in Phoenix 12.18+
- [11.23.2025: Repetitions for Manual Playground Invocations](/docs/phoenix/release-notes/11-2025/11-23-2025-repetitions-for-manual-playground-invocations.md): Available in Phoenix 12.17+
- [11.14.2025: Expanded Provider Support with OpenAI 5.1 + Gemini 3](/docs/phoenix/release-notes/11-2025/11-19-2025-expanded-provider-support-with-openai-5-1-+-gemini-3.md): Available in Phoenix 12.15+
- [11.12.2025: Updated Anthropic Model List](/docs/phoenix/release-notes/11-2025/11-12-2025-updated-anthropic-model-list.md): Available in Phoenix 12.15+
- [11.09.2025 OpenInference TypeScript 2.0](/docs/phoenix/release-notes/11-2025/11-09-2025-openinference-typescript-2-0.md): Added easy manual instrumentation with the same decorators, wrappers, and attribute helpers found in the Python openinference-instrumentation package.
- [11.07.2025: Timezone Preference](/docs/phoenix/release-notes/11-2025/11-07-2025-timezone-preference.md): Available in Phoenix 12.11+
- [11.05.2025: Metadata for Prompts](/docs/phoenix/release-notes/11-2025/11-05-2025-metadata-for-prompts.md): Available in Phoenix 12.10+
- [11.03.2025: Playground Dataset Label Display](/docs/phoenix/release-notes/11-2025/11-03-2025-playground-dataset-label-display.md): Available in Phoenix 12.10+
- [11.01.2025: Resume Experiments and Evaluations](/docs/phoenix/release-notes/11-2025/11-01-2025-resume-experiments-and-evaluations.md): Available in Phoenix 12.10+
- [10.30.2025: Metadata Support for Experiment Run Annotations](/docs/phoenix/release-notes/10-2025/10-30-2025-metadata-support-for-experiment-run-annotations.md): Available in Phoenix 12.9+
- [10.28.2025: Enable AWS IAM Auth for DB Configuration](/docs/phoenix/release-notes/10-2025/10-28-2025-enable-aws-iam-auth-for-db-configuration.md): Available in Phoenix 12.9+
- [10.26.2025: Add Split Edit Menu to Examples ䷖](/docs/phoenix/release-notes/10-2025/10-26-2025-add-split-edit-menu-to-examples.md): Available in Phoenix 12.8+
- [10.24.2025: Filter Prompts Page by Label](/docs/phoenix/release-notes/10-2025/10-24-2025-filter-prompts-page-by-label.md): Available in Phoenix 12.7+
- [10.20.2025: Splits ䷖](/docs/phoenix/release-notes/10-2025/10-20-2025-splits.md): Available in Phoenix 12.7+
- [10.18.2025: Filter Annotations in Compare Experiments Slideover](/docs/phoenix/release-notes/10-2025/10-18-2025-filter-annotations-in-compare-experiments-slideover.md): Available in Phoenix 12.7+
- [10.15.2025: Enhanced Filtering for Examples Table](/docs/phoenix/release-notes/10-2025/10-15-2025-enhanced-filtering-for-examples-table.md): Available in Phoenix 12.5+
- [10.13.2025: View Traces in Compare Experiments](/docs/phoenix/release-notes/10-2025/10-13-2025-view-traces-in-compare-experiments.md): Available in Phoenix 12.5+
- [10.10.2025: Viewer Role](/docs/phoenix/release-notes/10-2025/10-10-2025-viewer-role.md): Available in Phoenix 12.5+
- [10.08.2025: Dataset Labels](/docs/phoenix/release-notes/10-2025/10-08-2025-dataset-labels.md): Available in Phoenix 12.3+
- [10.06.2025: Paginate Compare Experiments](/docs/phoenix/release-notes/10-2025/10-06-2025-paginate-compare-experiments.md): Available in Phoenix 12.3+
- [10.05.2025: Load Prompt by Tag into Playground](/docs/phoenix/release-notes/10-2025/10-05-2025-load-prompt-by-tag-into-playground.md): Available in Phoenix 12.2+
- [10.03.2025: Prompt Version Editing in Playground](/docs/phoenix/release-notes/10-2025/10-03-2025-prompt-version-editing-in-playground.md): Available in Phoenix 12.2+
- [09.29.2025: Day 0 support for Claude Sonnet 4.5](/docs/phoenix/release-notes/09-2025/09-29-2025-day-0-support-for-claude-sonnet-4.5.md): Available in Phoenix 12.1+
- [09.27.2025: Dataset Splits](/docs/phoenix/release-notes/09-2025/09-27-2025-dataset-splits.md): Available in Phoenix 12.0+
- [09.26.2025: Session Annotations](/docs/phoenix/release-notes/09-2025/09-26-2025-session-annotations.md): Available in Phoenix 12.0+
- [09.25.2025: Repetitions](/docs/phoenix/release-notes/09-2025/09-25-2025-repetitions.md): Available in Phoenix 11.38+
- [09.24.2025: Custom HTTP headers for requests in Playground](/docs/phoenix/release-notes/09-2025/09-24-2025-custom-http-headers-for-requests-in-playground.md): Available in Phoenix 11.36+
- [09.23.2025: Repetitions in experiment compare slideover](/docs/phoenix/release-notes/09-2025/09-23-2025-repetitions-in-experiment-compare-slideover.md): Available in Phoenix 11.35+
- [09.22.2025: Helm configurable image registry & IPv6 support](/docs/phoenix/release-notes/09-2025/09-22-2025-helm-configurable-image-registry-and-ipv6-support.md): Available in Phoenix 11.35+
- [09.17.2025: Experiment compare details slideover in list view](/docs/phoenix/release-notes/09-2025/09-17-2025-experiment-compare-details-slideover-in-list-view.md): Available in Phoenix 11.34+
- [09.15.2025: Prompt Labels](/docs/phoenix/release-notes/09-2025/09-15-2025-prompt-labels.md): Available in Phoenix 11.33+
- [09.12.2025: Enable Paging in Experiment Compare Details](/docs/phoenix/release-notes/09-2025/09-12-2025-enable-paging-in-experiment-compare-details.md): Available in Phoenix 11.33+
- [09.08.2025: Experiment Annotation Popover in Detail View](/docs/phoenix/release-notes/09-2025/09-08-2025-experiment-annotation-popover-in-detail-view.md): Available in Phoenix 11.33+
- [09.04.2025: Experiment Lists Page Frontend Enhancements](/docs/phoenix/release-notes/09-2025/09-04-2025-experiment-lists-page-frontend-enhancements.md): Available in Phoenix 11.32+
- [09.03.2025: Add Methods to Log Document Annotations](/docs/phoenix/release-notes/09-2025/09-03-2025-add-methods-to-log-document-annotations.md): Available in Phoenix 11.31+
- [08.28.2025: New arize-phoenix-client Package](/docs/phoenix/release-notes/08-2025/08-28-2025-new-arize-phoenix-client-package.md): We’re excited to announce the release of arize-phoenix-client, a lightweight, fully-featured package for interacting with Phoenix. With this client, you no longer need to install arize-phoenix unless running a local instance.
- [08.22.2025: New Trace Timeline View](/docs/phoenix/release-notes/08-2025/08-22-2025-new-trace-timeline-view.md): Available in Phoenix 11.26+
- [08.20.2025: New Experiment and Annotation Quick Filters](/docs/phoenix/release-notes/08-2025/08-20-2025-new-experiment-and-annotation-quick-filters.md): Available in Phoenix 11.25+
- [08.15.2025: Enhance Experiment Comparison Views](/docs/phoenix/release-notes/08-2025/08-15-2025-enhance-experiment-comparison-views.md): Available in Phoenix 11.24+
- [08.14.2025: Trace Transfer for Long-Term Storage](/docs/phoenix/release-notes/08-2025/08-14-2025-trace-transfer-for-long-term-storage.md): Available in Phoenix 11.23+
- [08.12.2025: UI Design Overhauls](/docs/phoenix/release-notes/08-2025/08-12-2025-ui-design-overhauls.md): Available in Phoenix 11.22+
- [08.09.2025: Playground Support for GPT-5](/docs/phoenix/release-notes/08-2025/08-09-2025-playground-support-for-gpt-5.md): Available in Phoenix 11.21+
- [08.07.2025: Improved Error Handling in Prompt Playground](/docs/phoenix/release-notes/08-2025/08-07-2025-improved-error-handling-in-prompt-playground.md): Available in Phoenix 11.20+
- [08.06.2025: Expanded Search Capabilities](/docs/phoenix/release-notes/08-2025/08-06-2025-expanded-search-capabilities.md): Available in Phoenix 11.19+
- [08.05.2025: Claude Opus 4-1 Support](/docs/phoenix/release-notes/08-2025/08-05-2025-claude-opus-4-1-support.md): Available in Phoenix 11.19+
- [08.04.2025: Manual Project Creation & Trace Duplication](/docs/phoenix/release-notes/08-2025/08-04-2025-manual-project-creation-and-trace-duplication.md): Available in Phoenix 11.19+
- [08.03.2025: Delete Spans via REST API](/docs/phoenix/release-notes/08-2025/08-03-2025-delete-spans-via-rest-api.md): Available in Phoenix 11.19+
- [07.29.2025: Google GenAI evals](/docs/phoenix/release-notes/07-2025/07-29-2025-google-genai-evals.md): Google GenAI Evals We've added support for the GoogleGenAIModel in phoenix-evals, enabling direct access to Google's Gemini models through the…
- [07.25.2025: Project dashboards](/docs/phoenix/release-notes/07-2025/07-25-2025-project-dashboards.md): Available in Phoenix 11.12+
- [07.25.2025: Average metrics in experiment comparison table](/docs/phoenix/release-notes/07-2025/07-25-2025-average-metrics-in-experiment-comparison-table.md): Available in Phoenix 11.12+
- [07.21.2025: Project and trace management via GraphQL](/docs/phoenix/release-notes/07-2025/07-21-2025-project-and-trace-management-via-graphql.md): Available in Phoenix 11.9+
- [07.18.2025: OpenInference Java](/docs/phoenix/release-notes/07-2025/07-18-2025-openinference-java.md): OpenInference Java OpenInference Java is now available, providing a comprehensive solution for tracing AI applications using OpenTelemetry.
- [07.13.2025: Experiments module in phoenix-client](/docs/phoenix/release-notes/07-2025/07-13-2025-experiments-module-in-phoenix-client.md): Available in Phoenix 11.7+
- [07.09.2025: Baseline for experiment comparisons](/docs/phoenix/release-notes/07-2025/07-09-2025-baseline-for-experiment-comparisons.md): Available in Phoenix 11.4+
- [07.07.2025: Database disk usage monitor](/docs/phoenix/release-notes/07-2025/07-07-2025-databse-disk-usage-monitor.md): Available in Phoenix 11.5+
- [07.03.2025: Cost summaries in trace headers](/docs/phoenix/release-notes/07-2025/07-03-2025-cost-summaries-in-trace-headers.md): Available in Phoenix 11.4+
- [07.02.2025: Cursor MCP button](/docs/phoenix/release-notes/07-2025/07-02-2025-cursor-mcp-button.md): Available in Phoenix 11.3+
- [06.25.2025: Cost tracking](/docs/phoenix/release-notes/06-2025/06-25-2025-cost-tracking.md): Available in Phoenix 11.0+
- [06.25.2025: Amazon Bedrock support in playground](/docs/phoenix/release-notes/06-2025/06-25-2025-amazon-bedrock-support-in-playground.md): Available in Phoenix 10.15+
- [06.13.2025: Session filtering](/docs/phoenix/release-notes/06-2025/06-13-2025-session-filtering.md): Available in Phoenix 10.12+
- [06.13.2025: Enhanced span creation and logging](/docs/phoenix/release-notes/06-2025/06-13-2025-enhanced-span-creation-and-logging.md): Available in Phoenix 10.12+
- [06.12.2025: Dataset filtering](/docs/phoenix/release-notes/06-2025/06-12-2025-dataset-filtering.md): Available in Phoenix 10.11+
- [06.06.2025: Experiment progress graph](/docs/phoenix/release-notes/06-2025/06-06-2025-experiment-progress-graph.md): Available in Phoenix 10.9+
- [06.04.2025: Ollama support in playground](/docs/phoenix/release-notes/06-2025/06-04-2025-ollama-support-in-playground.md): Available in Phoenix 10.7+
- [06.03.2025: Deploy via helm](/docs/phoenix/release-notes/06-2025/06-03-2025-deploy-via-helm.md): Available in Phoenix 10.6+
- [05.30.2025: XAI and deepseek support in playground](/docs/phoenix/release-notes/05-2025/05-30-2025-xai-and-deepseek-support-in-playground.md): Available in Phoenix 10.5+
- [05.20.2025: Datasets and experiment evaluations in the JS client](/docs/phoenix/release-notes/05-2025/05-20-2025-datasets-and-experiment-evaluations-in-the-js-client.md): Datasets And Experiment Evaluations In The JS Client We've added a host of new methods to the JS client: getExperiment - allows you to retrieve an…
- [05.14.2025: Experiments in the JS client](/docs/phoenix/release-notes/05-2025/05-14-2025-experiments-in-the-js-client.md): Experiments In The JS Client You can now run Experiments using the Phoenix JS client!
- [05.09.2025: Annotations, data retention policies, hotkeys](/docs/phoenix/release-notes/05-2025/05-09-2025-annotations-data-retention-policies-hotkeys.md): Available in Phoenix 9.0.0+
- [05.05.2025: OpenInference Google GenAI instrumentation](/docs/phoenix/release-notes/05-2025/05-05-2025-openinference-google-genai-instrumentation.md): OpenInference Google GenAI Instrumentation We've added a Python auto-instrumentation library for the Google GenAI SDK.
- [04.30.2025: Span querying & data extraction for Phoenix client](/docs/phoenix/release-notes/04-2025/04-30-2025-span-querying-and-data-extraction-for-phoenix-client.md): Available in Phoenix 8.30+
- [04.28.2025: TLS support for Phoenix server](/docs/phoenix/release-notes/04-2025/04-28-2025-tls-support-for-phoenix-server.md): Available in Phoenix 8.29+
- [04.28.2025: Improved shutdown handling](/docs/phoenix/release-notes/04-2025/04-28-2025-improved-shutdown-handling.md): Available in Phoenix 8.28+
- [04.25.2025: Scroll selected span into view](/docs/phoenix/release-notes/04-2025/04-25-2025-scroll-selected-span-into-view.md): Available in Phoenix 8.27+
- [04.18.2025: Tracing for MCP client server applications](/docs/phoenix/release-notes/04-2025/04-18-2025-tracing-for-mcp-client-server-applications.md): Available in Phoenix 8.26+
- [04.16.2025: API key generation via API](/docs/phoenix/release-notes/04-2025/04-16-2025-api-key-generation-via-api.md): Available in Phoenix 8.26+
- [04.15.2025: Display tool call and result ids in span details](/docs/phoenix/release-notes/04-2025/04-15-2025-display-tool-call-and-result-ids-in-span-details.md): Available in Phoenix 8.25+
- [04.09.2025: Project management API enhancements](/docs/phoenix/release-notes/04-2025/04-09-2025-project-management-api-enhancements.md): Available in Phoenix 8.24+
- [04.09.2025: New REST API for projects with RBAC](/docs/phoenix/release-notes/04-2025/04-09-2025-new-rest-api-for-projects-with-rbac.md): Available in Phoenix 8.23+
- [04.03.2025: Phoenix client prompt tagging](/docs/phoenix/release-notes/04-2025/04-03-2025-phoenix-client-prompt-tagging.md): Available in Phoenix 8.22+
- [04.02.2025 improved span annotation editor](/docs/phoenix/release-notes/04-2025/04-02-2025-improved-span-annotation-editor.md): Available in Phoenix 8.21+
- [04.01.2025: Support for MCP span tool info in OpenAI agents SDK](/docs/phoenix/release-notes/04-2025/04-01-2025-support-for-mcp-span-tool-info-in-openai-agents-sdk.md): Available in Phoenix 8.20+
- [03.27.2025 span view improvements](/docs/phoenix/release-notes/03-2025/03-27-2025-span-view-improvements.md): Available in Phoenix 8.20+
- [03.24.2025: Tracing configuration tab](/docs/phoenix/release-notes/03-2025/03-24-2025-tracing-configuration-tab.md): Available in Phoenix 8.19+
- [03.21.2025: Environment variable based admin user configuration](/docs/phoenix/release-notes/03-2025/03-21-2025-environment-variable-based-admin-user-configuration.md): Available in Phoenix 8.17+
- [03.20.2025: Delete experiment from action menu](/docs/phoenix/release-notes/03-2025/03-20-2025-delete-experiment-from-action-menu.md): Available in Phoenix 8.19+
- [03.19.2025: Access to new integrations in projects](/docs/phoenix/release-notes/03-2025/03-19-2025-access-to-new-integrations-in-projects.md): Available in Phoenix 8.15+
- [03.18.2025: Resize span, trace, and session tables](/docs/phoenix/release-notes/03-2025/03-18-2025-resize-span-trace-and-session-tables.md): Available in Phoenix 8.14+
- [03.14.2025: OpenAI agents instrumentation](/docs/phoenix/release-notes/03-2025/03-14-2025-openai-agents-instrumentation.md): Available in Phoenix 8.13+
- [03.07.2025: Model config enhancements for prompts](/docs/phoenix/release-notes/03-2025/03-07-2025-model-config-enhancements-for-prompts.md): Available in Phoenix 8.11+
- [03.07.2025: New prompt playground, evals, and integration support](/docs/phoenix/release-notes/03-2025/03-07-2025-new-prompt-playground-evals-and-integration-support.md): Available in Phoenix 8.9+
- [03.06.2025: Project improvements](/docs/phoenix/release-notes/03-2025/03-06-2025-project-improvements.md): Available in Phoenix 8.5+
- [02.19.2025: Prompts](/docs/phoenix/release-notes/02-2025/02-19-2025-prompts.md): Available in Phoenix 8.0+
- [02.18.2025: One line instrumentation](/docs/phoenix/release-notes/02-2025/02-18-2025-one-line-instrumentation.md): Available in Phoenix 8.0+
- [01.18.2025: Automatic & manual span tracing](/docs/phoenix/release-notes/01-2025/01-18-2025-automatic-and-manual-span-tracing.md): Available in Phoenix 7.9+
- [12.09.2024: Sessions](/docs/phoenix/release-notes/2024/12-09-2024-sessions.md): Available in Phoenix 7.0+
- [11.18.2024: Prompt playground](/docs/phoenix/release-notes/2024/11-18-2024-prompt-playground.md): Available in Phoenix 6.0+
- [09.26.2024: Authentication & RBAC](/docs/phoenix/release-notes/2024/09-26-2024-authentication-and-rbac.md): Available in Phoenix 5.0+
- [07.18.2024: Guardrails AI integrations](/docs/phoenix/release-notes/2024/07-18-2024-guardrails-ai-integrations.md): Available in Phoenix 4.11+
- [07.03.2024: Datasets & experiments](/docs/phoenix/release-notes/2024/07-03-2024-datasets-and-experiments.md): Available in Phoenix 4.6+
- [07.02.2024: Function call evaluations](/docs/phoenix/release-notes/2024/07-02-2024-function-call-evaluations.md): Available in Phoenix 4.6+

## Other

- [PROMPT](/docs/PROMPT.md): Phoenix Tracing — Agent Setup Prompt You are helping a developer add Phoenix tracing to their application.
- [Quickstart: Datasets & experiments](/docs/phoenix/datasets-and-experiments/quickstart-datasets.md): Phoenix helps you run experiments over your AI and LLM applications to evaluate and iteratively improve their performance.
- [Arize Phoenix](/docs/phoenix/documentation/jp.md): なぜフェニックスなのか？ 完全にオープンソースでオープンに開発…
- [Arize Phoenix](/docs/phoenix/documentation/zh.md): 为什么选择Phoenix？ 完全开源，在开放环境中开发 Phoenix是一个开源平台，在设计时考虑了隐私和可定制性。Phoenix应用程序可以自行托管，完全在本地计算机或容器中运行，也可以通过我们网站上的免费云实例访问。\\…
- [Custom Task Evaluation](/docs/phoenix/evaluation/concepts-evals/building-your-own-evals.md): Customize Your Own Eval Templates The LLM Evals library is designed to support the building of any custom Eval templates.
- [Evaluating Multi-Agent Systems](/docs/phoenix/evaluation/concepts-evals/evaluating-multi-agent-systems.md): Evaluating multi-agent systems involves unique challenges compared to single-agent evaluations. This guide provides clear explanations of various architectures, strategies for effective evaluation, and additional considerations.
- [Eval Data Types](/docs/phoenix/evaluation/concepts-evals/evaluation-types.md): There are a multiple types of evaluations supported by the Phoenix Library. Each category of evaluation is categorized by its output type.
- [Evaluators](/docs/phoenix/evaluation/concepts-evals/evaluators.md): Definition and types of Evaluators. Score abstraction.
- [Input Mapping](/docs/phoenix/evaluation/concepts-evals/input-mapping.md): Evaluators are defined with a specific input schema, and the input payload is expected to take a certain shape.
- [Concepts: Evals](/docs/phoenix/evaluation/concepts-evals.md)
- [Quickstart: Evals](/docs/phoenix/evaluation/evals.md): Evaluate data from your LLM application using arize-phoenix-evals 3.0.
- [Prompt Formats](/docs/phoenix/evaluation/how-to-evals/configuring-the-llm/prompt-formats.md): Phoenix evaluators support multiple prompt formats, all compatible with supported models and providers.
- [Evaluator Traces](/docs/phoenix/evaluation/llm-evals/evaluator-traces.md): Phoenix Evals automatically traces all evaluation executions, providing complete transparency into how your evaluators make decisions.
- [Executors](/docs/phoenix/evaluation/llm-evals/executors.md): Phoenix Evals leverages executors that make the execution of evaluations many times faster.
- [Use Any LLM](/docs/phoenix/evaluation/llm-evals/use-any-llm.md)
- [Get Started: Evaluations](/docs/phoenix/evaluation/python-quickstart.md): Now that you have Phoenix up and running, and sent traces to your first project, the next step you can take is running evaluations of your Python…
- [Built-in Evaluators](/docs/phoenix/evaluation/server-evals/builtin-evaluators.md): Server-side code evaluators that label and score experiment outputs without requiring an LLM, API keys, or any installation.
- [TypeScript Quickstart](/docs/phoenix/evaluation/typescript-quickstart.md): The Phoenix TypeScript Evals package makes it easy to evaluate your application end-to-end by defining datasets, tasks, and evaluators that reveal…
- [Amazon Bedrock Evals](/docs/phoenix/integrations/llm-providers/amazon-bedrock/amazon-bedrock-evals.md): Configure and run Bedrock for evals
- [Anthropic Evals](/docs/phoenix/integrations/llm-providers/anthropic/anthropic-evals.md): Configure and run Anthropic for evals
- [Google ADK Tracing](/docs/phoenix/integrations/llm-providers/google-adk/google-adk-tracing.md): Instrument LLM calls made using the Google ADK Python SDK
- [Gemini Evals](/docs/phoenix/integrations/llm-providers/google-gen-ai/gemini-evals.md): Configure and run Gemini for evals
- [Evaluate CrewAI agents with Google gen AI evals](/docs/phoenix/integrations/llm-providers/google-gen-ai/google-gen-ai-evals-1.md): Evaluate multi-agent systems using Arize Phoenix, Google Evals, and CrewAI
- [Google Gen AI Evals](/docs/phoenix/integrations/llm-providers/google-gen-ai/google-gen-ai-evals.md): Using Google Gen AI with Phoenix Evals Need to install the extra dependency google-genai>=1.0.0 Use the LLM wrapper with provider="google" to access…
- [LiteLLM Evals](/docs/phoenix/integrations/llm-providers/litellm/litellm-evals.md): Configure and run LiteLLM for evals
- [MistralAI evals](/docs/phoenix/integrations/llm-providers/mistralai/mistralai-evals.md): Configure and run MistralAI for evals
- [OpenAI evals](/docs/phoenix/integrations/llm-providers/openai/openai-evals.md): Configure and run OpenAI for evals
- [VertexAI Evals](/docs/phoenix/integrations/llm-providers/vertexai/vertexai-evals.md): Configure and run VertexAI for evals
- [Model context protocol](/docs/phoenix/integrations/model-context-protocol.md): Anthropic's Model Context Protocol is a standard for connecting AI assistants to the systems where data lives, including content repositories, business tools, and development environments.
- [Launch Phoenix instructions](/docs/phoenix/integrations/sign-up-for-phoenix-sign-up.md): Install packages:
- [Concepts: Prompts](/docs/phoenix/prompt-engineering/concepts-prompts.md)
- [01.05.2026: Appended Messages for Playground Experiments](/docs/phoenix/release-notes/01-2026/01-05-2026-appended-messages-for-playground-experiments.md): Available in Phoenix 13.0+
- [01.21.2026 Phoenix CLI: Datasets, Experiments & Annotations for AI Coding Workflows](/docs/phoenix/release-notes/01-2026/01-21-2026-cli-datasets-experiments-annotations.md): Full documentation Install from npm Overview The Phoenix CLI now supports datasets, experiments, and annotations.
- [01.21.2026 Create Datasets from Traces with Span Associations](/docs/phoenix/release-notes/01-2026/01-21-2026-create-datasets-from-traces.md): Phoenix Client Documentation Overview Available in arize-phoenix-client 1.28.0+ (Python) and @arizeai/phoenix-client 2.0.0+ (TypeScript) Creating…
- [01.22.2026 CLI Prompt Commands: Pipe Prompts to AI Assistants](/docs/phoenix/release-notes/01-2026/01-22-2026-cli-prompt-commands.md): Full documentation Install from npm Overview Phoenix CLI now supports prompt introspection with two new commands: px prompts and px prompt.
- [Release Notes](/docs/phoenix/release-notes/02-2026/02-01-2026-tool-selection-and-tool-invocation-evaluators.md): Tool Selection and Tool Invocation Evaluators January 31, 2026 Available in arize-phoenix-evals 0.16.0+ (Python) and @arizeai/phoenix-evals 1.3.0+…
- [Release Notes](/docs/phoenix/release-notes/02-2026/02-10-2026-claude-opus-4-6-model-support.md): Claude Opus 4.6 Model Support February 9, 2026 Phoenix playground now supports Claude Opus 4.6, Anthropic's latest flagship model.
- [Release Notes](/docs/phoenix/release-notes/02-2026/02-11-2026-custom-providers.md): Custom Providers for Playground and Prompts February 11, 2026 Phoenix now supports custom providers for OpenAI, Azure OpenAI, Anthropic, AWS Bedrock,…
- [Dataset Evaluators](/docs/phoenix/release-notes/02-2026/02-12-2026-dataset-evaluators.md): Attach evaluators to datasets for automatic scoring during experiments.
- [OpenAI Responses API Type Support](/docs/phoenix/release-notes/02-2026/02-12-2026-openai-responses-api-support.md): Choose between Chat Completions and Responses API for OpenAI and Azure OpenAI.
- [2026](/docs/phoenix/release-notes/2026.md)
- [Annotation Config](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotationconfig.md)
- [Annotations](/docs/phoenix/sdk-api-reference/rest-api/api-reference/annotations.md): REST API methods for retrieving and deleting annotations on spans, traces, and sessions
- [Datasets](/docs/phoenix/sdk-api-reference/rest-api/api-reference/datasets.md): REST API methods for interacting with Phoenix datasets
- [Experiments](/docs/phoenix/sdk-api-reference/rest-api/api-reference/experiments.md): REST API methods for managing Phoenix experiments, runs, and evaluations
- [Projects](/docs/phoenix/sdk-api-reference/rest-api/api-reference/projects.md): REST API methods for interacting with Phoenix projects
- [Prompts](/docs/phoenix/sdk-api-reference/rest-api/api-reference/prompts.md): REST API methods for interacting with Phoenix prompts
- [Secrets](/docs/phoenix/sdk-api-reference/rest-api/api-reference/secrets.md)
- [Sessions](/docs/phoenix/sdk-api-reference/rest-api/api-reference/sessions.md): REST API methods for interacting with Phoenix sessions
- [Spans](/docs/phoenix/sdk-api-reference/rest-api/api-reference/spans.md): REST API methods for interacting with Phoenix spans
- [Traces](/docs/phoenix/sdk-api-reference/rest-api/api-reference/traces.md): REST API methods for interacting with Phoenix traces
- [Users](/docs/phoenix/sdk-api-reference/rest-api/api-reference/users.md)
- [API Reference](/docs/phoenix/sdk-api-reference/rest-api/api-reference.md): Use this page to navigate Phoenix REST endpoints by resource type.
- [Skill](/docs/phoenix/skill.md): Open-source AI observability platform for tracing, evaluating, and improving LLM applications with OpenTelemetry integration
- [Features: Tracing](/docs/phoenix/tracing/features-tracing.md): Tracing is a critical part of AI Observability and should be used both in production and development
- [Construct Shareable URLs](/docs/phoenix/tracing/how-to-tracing/advanced/constructing-urls.md): Link directly to projects, traces, spans, and sessions by human-readable identifiers
- [Evaluate RAG](/docs/phoenix/use-cases/rag-evaluation.md): Building a RAG pipeline and evaluating it with Phoenix Evals.
- [Agent assisted onboarding prereq](/docs/snippets/agent-assisted-onboarding-prereq.md): Don't have traces yet?
- [Agent assisted onboarding](/docs/snippets/agent-assisted-onboarding.md): Let your coding agent set this up. Works with Claude Code, Codex, Cursor, and OpenCode. In one terminal, start Phoenix and leave it running.
- [Connect to phoenix env](/docs/snippets/connect-to-phoenix-env.md): Point your code at the Phoenix instance you started.
- [Connect to phoenix python](/docs/snippets/connect-to-phoenix-python.md): Point your code at the Phoenix instance you started.
- [Launch phoenix docker](/docs/snippets/launch-phoenix-docker.md): Images are published to Docker Hub. See Docker for volumes, PostgreSQL, and other options.
- [Launch phoenix self host](/docs/snippets/launch-phoenix-self-host.md): Run Phoenix on your own infrastructure, backed by PostgreSQL so traces persist beyond a single process.
- [Launch phoenix terminal](/docs/snippets/launch-phoenix-terminal.md): No uv? pip install arize-phoenix && phoenix serve does the same thing. See Terminal setup for customization.
- [Phoenix auth note](/docs/snippets/phoenix-auth-note.md): Pointing at a deployment with authentication enabled?
- [Register tracer python](/docs/snippets/register-tracer-python.md): Connect your application to Phoenix with the register function: {from phoenix.otel import register configure the Phoenix tracer tracerprovider =…
- [Start phoenix](/docs/snippets/start-phoenix.md): Phoenix serves its UI and OTLP HTTP on port 6006, and OTLP gRPC on port 4317. For a local instance that's http://localhost:6006.