| Self-Hosted OSS | Arize Pro Plan | Arize Enterprise | |
|---|---|---|---|
| Logs (spans) | Your control | 10K | Custom volume |
| Storage | Your control | 10 GB | Custom volume |
Arize PX: Open-source LLM tracing & evaluation
Arize Phoenix (PX) gives AI engineers a seamless solution to evaluate, experiment, and optimize AI applications in development
Over 2.2 million downloads per month
Application Tracing for Total Visibility
Easily collect LLM app data with automatic instrumentation — or go manual for utmost control.
Interactive Prompt Playground
Leverage an eval library with unparalleled speed and ergonomics. Start with pre-built templates, easily customized to any task, or incorporate Human feedback.
Streamlined Evaluations and Annotations
Leverage an eval library with unparalleled speed and ergonomics. Start with pre-built templates, easily customized to any task, or incorporate Human feedback.
Dataset Clustering & Visualization
Uncover semantically similar questions, document chunks, and responses using embeddings to isolate poor performance.
Proudly Open Source
Built on top of OpenTelemetry, Phoenix is agnostic of vendor, framework, and language – granting you the flexibility you need in today’s generative landscape.
Which is right for you?
PX OSS
For developers who want to fully control the deployment
LLM / Generative
You control usage volume
Access via Cloud or Self-host
AX Pro
For small and establishing teams
LLM / Generative
10K spans
10GB storage
Up to 2 models or apps
AX Enterprise
For teams with global scale and advanced needs
LLM / Generative
ML Models (e.g. ranking)
Computer Vision
Custom spans volume
Custom storage volume
Custom volume of models/apps
| Self-Hosted OSS | Arize Pro Plan | Arize Enterprise | |
|---|---|---|---|
| LLM Tracing | |||
| Offline (code-based) evaluations | |||
| Offline evaluations — LLM as a Judge |
|||
| Online evaluations — LLM as a Judge |
|||
| Online evaluations — Python/TS code evaluator |
|||
| Sessions | |||
| AI Copilot | |||
| Prompt Playground | |||
| Prompt Hub | |||
| LLM Retrieval (RAG) analysis | |||
| Datasets | |||
| AI-powered dataset curation | |||
| Annotations | |||
| Human Feedback | |||
| Experiments | |||
| Datasets and Experiments | |||
| Embeddings analysis | |||
| Embeddings Similarity Search | |||
| Dashboards | |||
| Monitors | |||
| Alerting Integrations | |||
| Custom Metrics | |||
| Data Export | 5GB | ||
| API access |
| Self-Hosted OSS | Arize Pro Plan | Arize Enterprise | |
|---|---|---|---|
| History Lookback / Retention | Your control | 13 months (optional increase) |
|
| SSO | |||
| Social Sign-on | |||
| SAML | |||
| RBAC | |||
| Organizations & workspaces | |||
| On-premises deployment option | |||
| Compliance reporting | |||
| HIPAA reporting |
| Self-Hosted OSS | Arize Pro Plan | Arize Enterprise | |
|---|---|---|---|
| Support | Community | Community | Slack, email, phone |
| SLAs | Business hours | 3hr response window | |
| Customer success manager | |||
| Customer success engineering | Option |