Updated

Arize Phoenix

Source-available (Elastic License 2.0) LLM observability platform for tracing, evaluating, and debugging AI agents and applications, built on OpenTelemetry.

31 open-source projects declare Arize Phoenix in their code; no maker's live product we track uses it yet.

Self-hostableFree tierPaid from $50/mo

Profile checked

Works with AI agents:llms.txt
Ask your AI about this, with this page as the source:ChatGPT ↗Claude ↗Perplexity ↗

How makers use Arize Phoenix

What Arize Phoenix is used for

LLM Observability & Evals · Open-source platform →OpenTelemetry-based tracing and evals you can start locally in a notebook, then self-host or move to its cloud.Trade-off: Source-available under the Elastic License rather than a permissive open-source license.
Datadog50Langfuse35Helicone13LangSmith11Braintrust6Arize Phoenix0Makers' products on each, for LLM observability & evals

Arize Phoenix vs the alternatives makers chose

OptionMakers' productsFree forAt 10,000 usersOpen source
Datadog50 43%——No
Langfuse35 30%~5,000 users$29/mo · CoreYesCompare →
Helicone13 11%~1,000 users$116/mo · ProYes
LangSmith11 9%~520 users$475/mo · DeveloperNoCompare →
Braintrust6 5%——No
Weights & Biases3 3%——No
Arize Phoenix0 0%~2,500 users—No

For LLM observability & evals: how many of the 117 makers' products we know the LLM observability & evals of chose each option, and what each costs at 10,000 users.

Costs are the cheapest plan that fits, from each vendor's pricing page, at the usage our stack planner assumes for a product with that many users. Free for: the users its cheapest plan a business can use stays free for.

When to use Arize Phoenix, and when not

Pick Arize Phoenix if

Something else fits better if

  • You want tracing, prompts and evals in one tool you can self-host for free: Langfuse.
  • You want request logs and costs today by changing one base URL: Helicone.
  • Your app is built on LangChain or LangGraph: LangSmith.
  • Your main question is whether a prompt or model change made outputs better or worse: Braintrust.
From each decision's guide, written from what makers' products chose and why.

Phoenix is Arize AI's open platform for seeing and testing what an LLM app or agent does. It records each run as a trace (model calls, retrieval, tool use, custom logic), lets you score outputs with evaluators, and lets you replay and compare prompt or model changes on the same inputs before shipping them.

More about Arize Phoenix

It is a single containerized app (web UI, trace collector and a SQL database) that receives traces over OpenTelemetry's OTLP protocol. Instrumentation comes from Arize's OpenInference libraries, which auto-instrument providers such as OpenAI, Anthropic and Bedrock and frameworks such as LangChain, LlamaIndex, DSPy, Mastra and the Vercel AI SDK, with packages for Python, TypeScript and Java. You can start it locally with one command and open it on localhost.

Beyond tracing it includes LLM-as-a-judge and code evaluators, human annotation, versioned prompts with a playground and span replay, and datasets with experiments. Storage is SQLite by default and Postgres for production; Arize documents Docker, Kubernetes/Helm, AWS CloudFormation, Cloud Run, Railway and Render deployments, and every feature is available when self-hosted.

Each instance is a single tenant, so isolating teams means running more instances. You operate, scale and secure it yourself with community support; managed hosting, SLAs, monitors and SOC 2/HIPAA certifications belong to Arize's commercial AX platform instead.

How Arize Phoenix itself is built

29 tools, from its own code, website and Product Hunt page.

Who uses it

No maker's live product on record yet, and 31 open-source projects that declare it in their code.

We haven't found a maker's live product that uses Arize Phoenix yet — only the open-source projects below, which declare it in their code.

Dify
DifyPlatform for building LLM apps, agents and RAG workflows in a visual editor, hosted as D…LLM Observability & EvalsIn its code · source ↗
Paid
PaidBilling platform for AI agents that meters agent events into value-based units, prices a…LLM Observability & EvalsIn its code · source ↗
Ragas
RagasOpen-source Python library for evaluating LLM and RAG applications, with metrics such as…LLM Observability & EvalsIn its code · source ↗

Open source: a project that declares Arize Phoenix as a dependency in its public code — verifiable, but not necessarily a live product.

Alternatives to Arize Phoenix

All alternatives by situation →

Questions makers ask about Arize Phoenix

What are the alternatives to Arize Phoenix?

For LLM observability & evals, the other options makers' products we track chose most are Datadog (50), Langfuse (35), Helicone (13), LangSmith (11), Braintrust (6), Weights & Biases (3). Open source or self-hostable among them: Langfuse, Helicone, Weights & Biases. Each alternative, with the situation it fits, is on the Arize Phoenix alternatives page.

When should you use Arize Phoenix?

You already use OpenTelemetry or want vendor-neutral instrumentation (LLM observability & evals). Something else fits better if: you want tracing, prompts and evals in one tool you can self-host for free, pick Langfuse; you want request logs and costs today by changing one base URL, pick Helicone.

What do I need to run it myself?

Phoenix is one container with a UI, collector and database. It uses SQLite out of the box for local or single-user setups and PostgreSQL 14 or newer for production and multiple users. source ↗

Are any features held back from the self-hosted version?

No. Self-hosting on your own machines or cloud account is free under the Elastic License 2.0, and Arize states there are no feature gates. source ↗

Can several teammates share one instance?

Yes, once you enable authentication; it supports role-based access, API keys and OAuth2/SSO, and you invite users from Settings. For strict isolation between teams, Arize recommends separate instances, since one instance is one tenant. source ↗

How long does it keep traces?

Indefinitely by default. From version 9.0 you can set retention policies per project, by age or by trace count, and set the default with an environment variable at deploy time. source ↗

Is it SOC 2 or HIPAA certified?

Those certifications cover Arize's managed AX platform, not Phoenix. Because Phoenix runs on your infrastructure and data never reaches Arize, compliance depends on your own environment. source ↗

If I outgrow it, do I have to re-instrument my app?

No. Phoenix and Arize AX both use OpenTelemetry and OpenInference, so you point the same traces at the other destination, or at both. source ↗

Is Arize Phoenix free?

Yes — there is a free tier a small product can run on; paid use starts at $50/mo. source ↗

Is Arize Phoenix open source or self-hostable?

Not open source, and you can self-host it. source ↗

Can AI coding agents work with Arize Phoenix?

It serves an llms.txt docs index.

Who uses Arize Phoenix?

No maker's live product we track yet; 31 open-source projects declare it in their code. source ↗