# Arize AX Documentation ## Site Description Arize AX is an AI engineering platform that brings together observability, experimentation, and evaluation so teams can build and improve AI agents and applications with confidence. The platform provides comprehensive tools for tracing, evaluation, experimentation, prompt engineering, and production monitoring across the entire AI application lifecycle. Base URL: https://arize.com/docs --- ## Get Started Get started with Arize AX by exploring quickstart guides that walk you through core features like tracing, evaluation, experiments, and prompt engineering. These guides provide step-by-step instructions to help you set up and begin using the platform effectively. ### Overview & Quickstarts - https://arize.com/docs/ax - Arize AX overview: continuous improvement loop from observe and annotate through experiments, evals, and production monitoring - https://arize.com/docs/ax/core-workflows - The four core workflows (Instrument, Observe, Evaluate, Develop) and how they fit together - https://arize.com/docs/ax/get-started/get-started-tracing - Get started: set up tracing - https://arize.com/docs/ax/get-started/get-started-evaluations - Get started: write your first evaluation - https://arize.com/docs/ax/get-started/get-started-improve-your-agent - Get started: improve your agent with prompts, datasets, and experiments --- ## Alyx - AI Engineering Agent - https://arize.com/docs/ax/alyx - Alyx: Cursor / Claude Code-style AI agent in Arize AX; context (highlight / @ / type it); how to open; workflow tabs (Observe, Improve, Evaluate); integrations; data privacy - https://arize.com/docs/ax/set-up-with-ai-assistants - Set Up Arize with Skills (coding agents, MCP, paste-and-go) --- ## Improve The Improve section walks through the iteration loop for AI applications: collect a dataset of failures, craft and version the prompts you'll test, set up a controlled experiment, run it in Playground or in code, and automate the comparison in CI. ### Improve workflow - https://arize.com/docs/ax/improve/build-a-dataset - Turn traces, reviewer feedback, or synthetic examples into a reusable dataset; manage it and auto-add production rows - https://arize.com/docs/ax/improve/set-up-an-experiment - Trace-walk the failure, define a baseline, and pick Playground or code - https://arize.com/docs/ax/improve/experiment-in-playground - Fast path: iterate on prompts, models, and parameters in the UI and compare variants - https://arize.com/docs/ax/improve/experiment-in-code - Flexible path: log or run experiments from Python, TypeScript, or the CLI for pipelines, agents, and custom tasks - https://arize.com/docs/ax/improve/github-action-basics - Automate experiments as regression gates on every PR and deploy ### Datasets - https://arize.com/docs/ax/improve/build-a-dataset - Create, manage, append, export, and auto-curate datasets for repeatable experiments ### Experiments - https://arize.com/docs/ax/improve/set-up-an-experiment - Plan an experiment by choosing a dataset, baseline, variable, task, and evaluator - https://arize.com/docs/ax/improve/experiment-in-playground - Run and compare prompt experiments in the Playground - https://arize.com/docs/ax/improve/experiment-in-code - Run or log experiments from Python, TypeScript, CLI, or another runtime - https://arize.com/docs/ax/improve/experiment-in-playground#compare-experiments - Compare experiment results across runs - https://arize.com/docs/ax/improve/experiment-in-playground#classification-metrics - Compute classification metrics across experiment runs ### CI/CD with experiments - https://arize.com/docs/ax/improve/ci-cd-for-automated-experiments - CI/CD integration for automated experiments - https://arize.com/docs/ax/improve/github-action-basics - GitHub Actions integration - https://arize.com/docs/ax/improve/gitlab-ci-cd-basics - GitLab CI/CD integration - https://arize.com/docs/ax/improve/jenkins-integration - Jenkins integration - https://arize.com/docs/ax/improve/harness-ci-cd - Harness CI/CD integration ### Improve prompts - https://arize.com/docs/ax/improve/what-are-prompts - Prompt concepts and how prompts fit into the Improve workflow - https://arize.com/docs/ax/improve/create-a-prompt - Create prompts in Arize AX - https://arize.com/docs/ax/improve/test-a-prompt - Test prompts on datasets, spans, tools, and multimodal inputs - https://arize.com/docs/ax/improve/optimize-a-prompt - Optimize prompts with Alyx, Prompt Learning, and evaluation feedback - https://arize.com/docs/ax/improve/save-and-version-prompts - Save, version, compare, and deploy prompts with Prompt Hub - https://arize.com/docs/ax/improve/saving-and-managing-playground-views - Save and manage Playground views --- ## Instrument The Instrument section covers getting telemetry data flowing into Arize. Learn what traces are, set up tracing for your application, customize your spans with metadata and structured attributes, manage sessions, track costs, and configure advanced patterns like OpenTelemetry collector deployments, multi-service context propagation, and sampling. ### Getting started - https://arize.com/docs/ax/instrument/what-are-traces - Traces and spans: the data model Arize uses to observe LLM applications - https://arize.com/docs/ax/instrument/set-up-tracing - Set up tracing with arize-otel and OpenInference auto-instrumentors - https://arize.com/docs/ax/instrument/manual-instrumentation - Manually create and customize spans in your own code ### Customizing traces - https://arize.com/docs/ax/instrument/customize-your-traces - Add inputs/outputs, metadata, attributes, events, and structured tags to spans - https://arize.com/docs/ax/instrument/set-up-sessions - Group related spans into sessions and track users - https://arize.com/docs/ax/instrument/combining-auto-and-manual - Layer manual spans on top of auto-instrumentation without duplication - https://arize.com/docs/ax/instrument/agent-trajectory - Capture agent trajectory and tool-calling traces - https://arize.com/docs/ax/instrument/track-costs - Cost tracking across providers (input/output tokens, cached tokens, custom costs) ### Advanced patterns - https://arize.com/docs/ax/instrument/configure-your-tracer - Configure the OTel tracer: resource attributes, processors, multi-space routing with register_with_routing - https://arize.com/docs/ax/instrument/mask-and-redact-data - PII masking, attribute filtering, and sensitive-data redaction before spans leave your app - https://arize.com/docs/ax/instrument/advanced-patterns - OTel Collector deployment (agent/gateway/hybrid), shared-collector pattern with arize-space-id routing, context propagation, custom sampling, health-check patterns --- ## Evaluate The Evaluate section is organized as a workflow: human review and annotations, creating LLM-as-a-judge and code evaluators (including UI tutorials), running online evals on traces, running offline evals on experiments, viewing results and dashboards, and tracking eval cost. ### Workflow pages - https://arize.com/docs/ax/evaluate/human-review - Annotation configs, labeling workflows, and annotating spans (UI, Alyx, Skills, code) for ground truth - https://arize.com/docs/ax/evaluate/labeling-queues - Labeling Queues: focused review workflows and ground truth datasets for evals - https://arize.com/docs/ax/evaluate/create-evaluators - Create LLM-as-a-judge and code evaluators (Alyx, UI, Skills tabs), Evaluator Hub, practices, and further reading - https://arize.com/docs/ax/evaluate/run-evals-on-traces - Run online evals on traces: tasks (Skills, Alyx, UI, By Code), sampling, filters, viewing results - https://arize.com/docs/ax/evaluate/run-evals-on-experiments - Run offline evals on experiments: evaluations on datasets and experiment runs for CI/CD and regression checks - https://arize.com/docs/ax/evaluate/evals-overview - Where to find eval results, monitor quality over time, and track or reduce evaluation spend - https://arize.com/docs/ax/evaluate/align-evals-to-human-feedback - Validate evals against human labels (Prompt Playground, agreement checks) before scaling ### Additional evaluation reference - https://arize.com/docs/ax/evaluate/evaluators/code-evaluations - Code evaluators: deterministic Python sandbox for objective checks - https://arize.com/docs/ax/evaluate/evaluators/retrieval-evaluation - Retrieval (RAG) evaluation - https://arize.com/docs/ax/cookbooks/evaluation/retrieval-evaluation - Retrieval (RAG) evaluation cookbook --- ## Observe The Observe section covers exploring and understanding your traces once they're flowing into Arize. View and filter live traces, get agent insights, take action on what you see (annotate, curate datasets, fix prompts, send to labeling queues), organize work into projects with custom metrics, build dashboards, and configure production monitors. ### Tracing - https://arize.com/docs/ax/observe/tracing/view-and-manage-traces - Viewing, filtering, and managing live traces - https://arize.com/docs/ax/observe/tracing/saved-views - Saved views for the Tracing table (filters, columns, sort, time range) - https://arize.com/docs/ax/observe/tracing/agents - Agent visualizations and analysis ### Agent insights - https://arize.com/docs/ax/observe/agent-insights - Aggregated insights across agents: common failure patterns, anomaly detection, convergence ### Take action - https://arize.com/docs/ax/observe/take-action - Take action on what you see in traces - https://arize.com/docs/ax/observe/take-action/annotate-traces - Annotate traces with labels for ground truth and evaluation - https://arize.com/docs/ax/observe/take-action/curate-dataset - Curate datasets from traces for experiments and evaluation - https://arize.com/docs/ax/observe/take-action/fix-a-prompt - Jump from a problematic trace into the Prompt Playground to iterate - https://arize.com/docs/ax/observe/take-action/labeling-queue - Send spans to labeling queues for focused human review ### Projects - https://arize.com/docs/ax/observe/projects/custom-metrics-api - Custom metrics API — common patterns, worked examples, and full AQL syntax reference ### Dashboards - https://arize.com/docs/ax/observe/dashboards - Dashboards overview with widgets, templates, and token-counting examples ### Production Monitoring - https://arize.com/docs/ax/observe/production-monitoring - Production monitoring overview with monitor configuration and alerting integrations (Slack, OpsGenie, PagerDuty) --- ## Integrations Arize AX supports integrations with over 30+ LLM providers, agent frameworks, platforms, and tools for seamless instrumentation. These integrations enable automatic tracing, evaluation capabilities, and observability across your entire AI stack without manual instrumentation. ### Overview - https://arize.com/docs/ax/integrations - Integrations overview and types ### LLM Provider Integrations - https://arize.com/docs/ax/integrations/llm-providers/openai - OpenAI integration (tracing, evals, Agents SDK tracing, Node.js SDK) - https://arize.com/docs/ax/integrations/llm-providers/anthropic - Anthropic integration (tracing, evals) - https://arize.com/docs/ax/integrations/llm-providers/google-gen-ai - Google Generative AI integration (tracing, Gemini evals) - https://arize.com/docs/ax/integrations/llm-providers/amazon-bedrock - Amazon Bedrock integration (tracing, evals, agents tracing) - https://arize.com/docs/ax/integrations/llm-providers/vertexai - Vertex AI integration (tracing, evals) - https://arize.com/docs/ax/integrations/llm-providers/mistralai - Mistral AI integration (tracing, evals) - https://arize.com/docs/ax/integrations/llm-providers/groq - Groq integration (tracing) - https://arize.com/docs/ax/integrations/llm-providers/openrouter - OpenRouter integration (tracing) - https://arize.com/docs/ax/integrations/llm-providers/litellm - LiteLLM integration (tracing, evals) - https://arize.com/docs/ax/integrations/llm-providers/llama - Llama/Ollama integration (tracing) ### Python Agent Framework Integrations - https://arize.com/docs/ax/integrations/python-agent-frameworks/langchain - LangChain integration (tracing, JS support) - https://arize.com/docs/ax/integrations/python-agent-frameworks/langgraph - LangGraph integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/llamaindex - LlamaIndex integration (tracing, workflows tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/autogen - AutoGen integration (tracing, AgentChat tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/crewai - CrewAI integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/dspy - DSPy integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/haystack - Haystack integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/instructor - Instructor integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/guardrails-ai - Guardrails AI integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/pydantic - Pydantic AI integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/semantic-kernel - Semantic Kernel integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/together-ai - Together AI integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/portkey - Portkey integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/google-adk - Google ADK integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/agno - Agno integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/beeai - BeeAI integration (Python tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/hugging-face-smolagents - Hugging Face smolagents integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/model-context-protocol - Model Context Protocol (MCP) integration (tracing) - https://arize.com/docs/ax/integrations/python-agent-frameworks/aws-strands - AWS Strands integration (tracing, Bedrock AgentCore) ### TypeScript/JavaScript Agent Framework Integrations - https://arize.com/docs/ax/integrations/ts-js-agent-frameworks/langchain - LangChain JS/TS integration (tracing) - https://arize.com/docs/ax/integrations/ts-js-agent-frameworks/mastra - Mastra integration (tracing) - https://arize.com/docs/ax/integrations/ts-js-agent-frameworks/vercel - Vercel AI SDK integration (tracing) - https://arize.com/docs/ax/integrations/ts-js-agent-frameworks/beeai - BeeAI JS integration (tracing) ### Java Integrations - https://arize.com/docs/ax/integrations/java/langchain4j - LangChain4j integration (tracing) - https://arize.com/docs/ax/integrations/java/spring-ai - Spring AI integration (tracing) - https://arize.com/docs/ax/integrations/java/arconia - Arconia integration (tracing) ### Coding Agent Integrations - https://arize.com/docs/ax/integrations/platforms/claude-code - Claude Code integration overview - https://arize.com/docs/ax/integrations/platforms/claude-code/claude-code-tracing - Claude Code tracing with Arize AX - https://arize.com/docs/ax/integrations/platforms/codex - Codex integration overview - https://arize.com/docs/ax/integrations/platforms/codex/codex-tracing - Codex CLI tracing with Arize AX - https://arize.com/docs/ax/integrations/platforms/cursor - Cursor integration overview - https://arize.com/docs/ax/integrations/platforms/cursor/cursor-tracing - Cursor tracing with Arize AX using Cursor hooks ### Platform Integrations - https://arize.com/docs/ax/integrations/platforms/langflow - LangFlow integration (tracing) - https://arize.com/docs/ax/integrations/platforms/flowise - Flowise integration (tracing) - https://arize.com/docs/ax/integrations/platforms/dify - Dify integration (tracing) - https://arize.com/docs/ax/integrations/platforms/prompt-flow - Prompt Flow integration (tracing) ### OpenTelemetry Integrations - https://arize.com/docs/ax/integrations/opentelemetry/overview - OpenTelemetry overview - https://arize.com/docs/ax/integrations/opentelemetry/opentelemetry-arize-otel - OpenTelemetry Arize OTel integration - https://arize.com/docs/ax/integrations/opentelemetry/openlit - OpenLIT integration - https://arize.com/docs/ax/integrations/opentelemetry/openllmetry - OpenLLMetry integration - https://arize.com/docs/ax/integrations/opentelemetry/traceloop-sdk - Traceloop SDK integration ### Other Evaluation Integrations - https://arize.com/docs/ax/integrations/evaluation-integrations/ragas - RAGAS integration (NVIDIA RAG metrics) - https://arize.com/docs/ax/integrations/evaluation-integrations/microsoft - Microsoft evaluation integration ### Vector Database Integrations - https://arize.com/docs/ax/integrations/vector-databases - Vector database integrations overview --- ## API Reference The API Reference section provides comprehensive documentation for all Arize AX APIs, including REST and GraphQL endpoints. It also includes SDK documentation for Python and TypeScript, with tutorials and migration guides to help you integrate programmatically. ### REST API - https://arize.com/docs/ax/rest-reference/overview - Arize REST API overview - https://arize.com/docs/ax/rest-reference/download-openapi-spec - Download OpenAPI specification - https://api.arize.com/v2/spec.yaml - OpenAPI specification (YAML) ### GraphQL API - https://arize.com/docs/ax/graphql-reference - GraphQL API reference overview - https://arize.com/docs/ax/graphql-reference/overview/getting-started-with-graphql - Getting started with GraphQL - https://arize.com/docs/ax/graphql-reference/overview/how-to-use-graphql - How to use GraphQL - https://arize.com/docs/ax/graphql-reference/overview/how-to-use-graphql/forming-calls - Forming GraphQL calls - https://arize.com/docs/ax/graphql-reference/overview/how-to-use-graphql/using-global-node-ids - Using global node IDs - https://arize.com/docs/ax/graphql-reference/overview/how-to-use-graphql/querying-nested-data - Querying nested data - https://arize.com/docs/ax/graphql-reference/overview/how-to-use-graphql/notebook-examples - Notebook examples - https://arize.com/docs/ax/graphql-reference/overview/how-to-use-graphql/mutations - GraphQL mutations - https://arize.com/docs/ax/graphql-reference/overview/resource-limitations - Resource limitations - https://arize.com/docs/ax/graphql-reference/apis/admin-api - Admin API - https://arize.com/docs/ax/graphql-reference/apis/annotations-api - Annotations API - https://arize.com/docs/ax/graphql-reference/apis/custom-metrics-api - Custom metrics API - https://arize.com/docs/ax/graphql-reference/apis/dashboards-api - Dashboards API - https://arize.com/docs/ax/graphql-reference/apis/file-importer-api - File importer API - https://arize.com/docs/ax/graphql-reference/apis/metrics-api - Metrics API - https://arize.com/docs/ax/graphql-reference/apis/models-api - Models API - https://arize.com/docs/ax/graphql-reference/apis/monitors-api - Monitors API - https://arize.com/docs/ax/graphql-reference/apis/online-tasks-api - Online tasks API - https://arize.com/docs/ax/graphql-reference/apis/table-importer-api - Table importer API ### Python SDK (API Clients) - https://arize.com/docs/api-clients/python/overview - Python SDK overview - https://arize.com/docs/api-clients/python/version-8/overview - Python SDK v8 overview (pre-release) - https://arize.com/docs/api-clients/python/version-8/tracing - Tracing with Python SDK v8 - https://arize.com/docs/api-clients/python/version-8/embeddings - Embeddings with Python SDK v8 - https://arize.com/docs/api-clients/python/version-8/client-resources/annotation-configs - Annotation configs client [Beta] - https://arize.com/docs/api-clients/python/version-8/client-resources/datasets - Datasets client [Beta] - https://arize.com/docs/api-clients/python/version-8/client-resources/experiments - Experiments client [Beta] - https://arize.com/docs/api-clients/python/version-8/client-resources/projects - Projects client [Beta] - https://arize.com/docs/api-clients/python/version-8/client-resources/spans - Spans client [Alpha] - https://arize.com/docs/api-clients/python/version-8/client-resources/spaces - Spaces client [Alpha] - https://arize.com/docs/api-clients/python/version-8/client-resources/api-keys - API keys client [Alpha] - https://arize.com/docs/api-clients/python/version-8/client-resources/ai-integrations - AI integrations client [Alpha] - https://arize.com/docs/api-clients/python/version-8/client-resources/ml-models - ML models client - https://arize.com/docs/api-clients/python/version-8/migration/index - Migration guide from v7 - https://arize.com/docs/api-clients/python/version-8/migration/datasets-client - Datasets client migration - https://arize.com/docs/api-clients/python/version-8/migration/experiments-client - Experiments client migration - https://arize.com/docs/api-clients/python/version-8/migration/exporter-client - Exporter client migration - https://arize.com/docs/api-clients/python/version-8/migration/pandas-client - Pandas client migration - https://arize.com/docs/api-clients/python/version-8/migration/stream-client - Stream client migration - https://arize.com/docs/api-clients/python/version-8/tutorial/get-started-tracing-with-arize-sdk - Get started with tracing tutorial - https://arize.com/docs/api-clients/python/version-8/tutorial/get-started-datasets-experiments-with-arize-sdk - Get started with datasets and experiments tutorial - https://arize.com/docs/api-clients/python/version-8/tutorial/get-started-evaluations-with-arize-sdk - Get started with evaluations tutorial - https://arize.com/docs/api-clients/python/version-7/overview - Python SDK v7 overview ### TypeScript SDK (API Clients) - https://arize.com/docs/api-clients/typescript/version-1/overview - TypeScript SDK v1 overview [Beta] - https://arize.com/docs/api-clients/typescript/version-1/client-resources/annotation-configs - Annotation configs client [Beta] - https://arize.com/docs/api-clients/typescript/version-1/client-resources/datasets - Datasets client [Beta] - https://arize.com/docs/api-clients/typescript/version-1/client-resources/experiments - Experiments client [Beta] - https://arize.com/docs/api-clients/typescript/version-1/client-resources/projects - Projects client [Beta] --- ## Security & Settings The Security & Settings section covers all security, compliance, and configuration options for your Arize AX instance. This includes API key management, SSO and RBAC setup, compliance features, cost tracking, guardrails, and integration playgrounds for testing LLM connections. ### API Keys & Authentication - https://arize.com/docs/ax/security-and-settings/api-keys - API keys management - https://arize.com/docs/ax/security-and-settings/service-keys - Service keys ### Security Features - https://arize.com/docs/ax/security-and-settings/llm-security/guardrails - Guardrails for LLM security - https://arize.com/docs/ax/security-and-settings/llm-security/llm-red-teaming - LLM red teaming - https://arize.com/docs/ax/security-and-settings/whitelisting - Whitelisting configuration - https://arize.com/docs/ax/security-and-settings/arize-private-connect - Arize Private Connect ### SSO & RBAC - https://arize.com/docs/ax/security-and-settings/sso-and-rbac - SSO and RBAC overview - https://arize.com/docs/ax/security-and-settings/sso-and-rbac/saml-configuration - SAML configuration - https://arize.com/docs/ax/security-and-settings/sso-and-rbac/setting-up-sso-with-okta - Setting up SSO with Okta - https://arize.com/docs/ax/security-and-settings/sso-and-rbac/rbac-rest-api - RBAC REST API: manage roles, role bindings, and resource restrictions programmatically - https://arize.com/docs/ax/security-and-settings/sso-and-rbac/custom-roles - Custom roles: create fine-grained permission sets - https://arize.com/docs/ax/security-and-settings/sso-and-rbac/project-restrictions - Project-level access restrictions ### Compliance - https://arize.com/docs/ax/security-and-settings/compliance - Compliance overview - https://arize.com/docs/ax/security-and-settings/compliance/arize-audit-log - Audit log - https://arize.com/docs/ax/security-and-settings/compliance/delete-traces-with-sensitive-data - Deleting traces with sensitive data ### Cost Tracking - https://arize.com/docs/ax/instrument/track-costs - Cost tracking configuration - https://arize.com/docs/ax/security-and-settings/space-rate-limiting - Per-space monthly ingestion limits in Organization settings; enterprise, opt-in; admins rebalance caps; contract total via Arize/support team ### Data Fabric - https://arize.com/docs/ax/security-and-settings/data-fabric - Data fabric overview ### Tags - https://arize.com/docs/ax/security-and-settings/tags - Tags management ### Integrations Playground - https://arize.com/docs/ax/security-and-settings/integrations-playground/openai - OpenAI integration playground - https://arize.com/docs/ax/security-and-settings/integrations-playground/azure-openai - Azure OpenAI integration playground - https://arize.com/docs/ax/security-and-settings/integrations-playground/vertexai - Vertex AI integration playground - https://arize.com/docs/ax/security-and-settings/integrations-playground/aws-bedrock - AWS Bedrock integration playground - https://arize.com/docs/ax/security-and-settings/integrations-playground/custom-llm-models - Custom LLM models playground ### Phoenix Migration - https://arize.com/docs/ax/security-and-settings/send-traces-from-phoenix-greater-than-arize - Sending traces from Phoenix to Arize --- ## Self-Hosting The Self-Hosting section provides complete documentation for deploying Arize AX on-premise or in your own cloud infrastructure. It includes installation guides for AWS, Azure, GCP, and OpenShift, along with configuration options for ingress, SAML, and SDK usage in self-hosted environments. ### Overview - https://arize.com/docs/ax/selfhosting - Self-hosting overview - https://arize.com/docs/ax/selfhosting/architecture - Architecture - https://arize.com/docs/ax/selfhosting/getting-started/overview - Getting started (installation flow) - https://arize.com/docs/ax/selfhosting/getting-started/prerequisites - Prerequisites - https://arize.com/docs/ax/selfhosting/installation/installation-on-aws - Installation on AWS - https://arize.com/docs/ax/selfhosting/installation/installation-on-azure - Installation on Azure - https://arize.com/docs/ax/selfhosting/installation/installation-on-gcp - Installation on GCP - https://arize.com/docs/ax/selfhosting/installation/installation-on-openshift - Installation on OpenShift - https://arize.com/docs/ax/selfhosting/installation/installation-on-single-host - Installation on Single Host - https://arize.com/docs/ax/selfhosting/installation/configuring-ingress-endpoints - Configuring ingress endpoints - https://arize.com/docs/ax/selfhosting/installation/configuring-saml - Configuring SAML - https://arize.com/docs/ax/selfhosting/guides/integrations - Integrations - https://arize.com/docs/ax/selfhosting/guides/sdk-usage - On-premise SDK usage - https://arize.com/docs/ax/selfhosting/guides/releases - On-premise releases --- ## Machine Learning The Machine Learning section covers ML observability for traditional machine learning models, including data upload, monitoring, drift detection, explainability, and performance analysis. It also includes computer vision observability, use cases for different model types, integrations with ML platforms, and comprehensive API references. ### Machine Learning Observability - https://arize.com/docs/ax/machine-learning/machine-learning - Machine learning observability overview - https://arize.com/docs/ax/machine-learning/machine-learning/quickstart - ML observability quickstart - https://arize.com/docs/ax/machine-learning/machine-learning/concepts-ml - ML concepts ### How To: ML - https://arize.com/docs/ax/machine-learning/machine-learning/how-to-ml - How-to guides for ML observability - https://arize.com/docs/ax/machine-learning/machine-learning/how-to-ml/upload-data-to-arize - Uploading data to Arize - https://arize.com/docs/ax/machine-learning/machine-learning/how-to-ml/monitors - Setting up monitors - https://arize.com/docs/ax/machine-learning/machine-learning/how-to-ml/dashboards - Creating dashboards - https://arize.com/docs/ax/machine-learning/machine-learning/how-to-ml/drift-tracing - Drift tracing - https://arize.com/docs/ax/machine-learning/machine-learning/how-to-ml/performance-tracing - Performance tracing - https://arize.com/docs/ax/machine-learning/machine-learning/how-to-ml/explainability - Model explainability - https://arize.com/docs/ax/machine-learning/machine-learning/how-to-ml/custom-metrics-api - Custom metrics API for ML ### ML Use Cases - https://arize.com/docs/ax/machine-learning/machine-learning/use-cases-ml - ML use cases overview ### ML Integrations - https://arize.com/docs/ax/machine-learning/machine-learning/integrations-ml - ML integrations overview ### ML API Reference - https://arize.com/docs/ax/machine-learning/machine-learning/api-reference-ml - ML API reference overview - https://arize.com/docs/ax/machine-learning/machine-learning/api-reference-ml/python-sdk - Python SDK reference - https://arize.com/docs/ax/machine-learning/machine-learning/api-reference-ml/java-sdk - Java SDK reference - https://arize.com/docs/ax/machine-learning/machine-learning/api-reference-ml/r-sdk - R SDK reference - https://arize.com/docs/ax/machine-learning/machine-learning/api-reference-ml/grpc-api - gRPC API reference ### Computer Vision - https://arize.com/docs/ax/machine-learning/computer-vision - Computer vision observability overview --- ## Cookbooks The Cookbooks section contains practical tutorials and examples for building AI applications with Arize AX, covering agents, evaluations, experiments, and prompt optimization. These step-by-step guides demonstrate real-world workflows and best practices for common AI engineering tasks. ### Overview - https://arize.com/docs/ax/cookbooks - Cookbooks and tutorials overview ### Agent Cookbooks - https://arize.com/docs/ax/cookbooks/agent-workflow-patterns - Agent workflow patterns - https://arize.com/docs/ax/cookbooks/agents/tracing-and-evaluating-agents - Tracing and evaluating agents - https://arize.com/docs/ax/cookbooks/agents/arize-+-mosaic-ai-agent-framework - Online evals and monitoring for agents in production - https://arize.com/docs/ax/cookbooks/agents/evaluating-agentic-rag-using-arize-and-couchbase - Evaluating agentic RAG with Couchbase - https://arize.com/docs/ax/cookbooks/agents/evaluating-and-improving-ai-agents-at-scale-with-microsoft-foundry - Evaluating and improving AI agents at scale with Microsoft Foundry - https://arize.com/docs/ax/cookbooks/agents/foundry-red-team - Foundry red team - https://arize.com/docs/ax/cookbooks/agents/openai-agents-cookbook - OpenAI agents cookbook - https://arize.com/docs/ax/cookbooks/agents/ragas-agents-cookbook - RAGAS agents cookbook - https://arize.com/docs/ax/cookbooks/agents/trace-evaluate-browser-use-agent-with-l-lama4 - Trace and evaluate browser use agent with Llama 4 - https://arize.com/docs/ax/cookbooks/agents/tracing-a-langgraph-application-with-agent-engine-in-vertex-ai - Tracing a LangGraph application with Agent Engine in Vertex AI - https://arize.com/docs/ax/cookbooks/agents/tracing-a-routing-agent - Tracing a routing agent - https://arize.com/docs/ax/cookbooks/agents/tracing-a2a-agent - Tracing A2A agent ### Evaluation Cookbooks - https://arize.com/docs/ax/cookbooks/evaluation/evaluation - Evaluation cookbooks overview - https://arize.com/docs/ax/cookbooks/evaluation/evaluating-rag - Evaluating RAG systems - https://arize.com/docs/ax/cookbooks/evaluation/gemini-audio-evals - Gemini audio evaluations - https://arize.com/docs/ax/cookbooks/evaluation/llamaindex-evals - LlamaIndex evaluations - https://arize.com/docs/ax/cookbooks/evaluation/session-level-evaluations-for-an-ai-tutor - Session-level evaluations for an AI tutor - https://arize.com/docs/ax/cookbooks/evaluation/trace-level-evaluations-for-a-recommendation-agent - Trace-level evaluations for a recommendation agent - https://arize.com/docs/ax/cookbooks/evaluation/tracing-and-evaluating-audio - Tracing and evaluating audio ### Experiment Cookbooks - https://arize.com/docs/ax/cookbooks/experiments/model-comparison-for-an-email-text-extraction-service - Model comparison for email text extraction - https://arize.com/docs/ax/cookbooks/experiments/summarization - Summarization experiments - https://arize.com/docs/ax/cookbooks/experiments/text2sql - Text-to-SQL experiments ### Prompt Learning Cookbooks - https://arize.com/docs/ax/cookbooks/prompt-learning/optimizing-your-eval-prompts - Optimizing your eval prompts - https://arize.com/docs/ax/cookbooks/prompt-learning/improving-structured-output-generation-with-prompt-learning - Improving structured output generation - https://arize.com/docs/ax/cookbooks/prompt-learning/optimizing-coding-agent-prompts-for-execution - Optimizing coding agent prompts for execution - https://arize.com/docs/ax/cookbooks/prompt-learning/optimizing-coding-agent-prompts-for-planning - Optimizing coding agent prompts for planning ### AI Engineering Workflows - https://arize.com/docs/ax/cookbooks/ai-engineering-workflows/guardrails - Guardrails for realtime detection ### Human-in-the-Loop Workflows - https://arize.com/docs/ax/cookbooks/human-in-the-loop-workflows-annotations/creating-a-custom-llm-evaluator-with-a-benchmark-dataset - Creating a custom LLM evaluator with a benchmark dataset ### Tracing Integrations - https://arize.com/docs/ax/cookbooks/tracing-integrations - Tracing integrations cookbook --- ## Additional Resources Find community support, blog posts, and related documentation to help you get the most out of Arize AX. Connect with other developers, stay updated on new features, and explore our open-source Phoenix project for additional observability tools. - Community Slack: https://arize-ai.slack.com/ssb/redirect#/shared-invite/email - Blog: https://arize.com/blog/ - Phoenix OSS Documentation: http://docs.arize.com/phoenix