From 8c1b2fcbc2a7d96181b5b07e71550d459752d44c Mon Sep 17 00:00:00 2001 From: Dat Ngo <97070834+arizedatngo@users.noreply.github.com> Date: Mon, 14 Sep 2026 11:37:56 +0200 Subject: [PATCH] docs: restore Arize evaluation resources removed by sync --- guides/community/arize.mdx | 4 +++- guides/features/broadcast/arize.mdx | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/guides/community/arize.mdx b/guides/community/arize.mdx index 0583c3f8..03de1064 100644 --- a/guides/community/arize.mdx +++ b/guides/community/arize.mdx @@ -20,7 +20,9 @@ import { API_KEY_REF } from '/snippets/exports/constants.mdx'; {/* */} ## Using Arize AX -[Arize AX](https://arize.com/) provides observability and tracing for LLM applications. Since OpenRouter uses the OpenAI API schema, you can use Arize's OpenInference auto-instrumentation with the OpenAI SDK to automatically trace and monitor your OpenRouter API calls. +[Arize AX](https://arize.com/docs/ax) is Arize AI's enterprise observability and evaluation product. Use OpenInference instrumentation with the OpenAI SDK to trace OpenRouter requests in AX, available as managed cloud or an enterprise self-hosted deployment. + +For an open-source workflow, use [Arize Phoenix](https://arize.com/docs/phoenix). To turn captured requests into quality checks, see Arize's [LLM evaluation guide](https://arize.com/resources/llm-evaluation/) and [agent evaluation guide](https://arize.com/guides/ai-agent-handbook/agent-evaluation/). ### Installation diff --git a/guides/features/broadcast/arize.mdx b/guides/features/broadcast/arize.mdx index 4421e4bc..fb3b6c1c 100644 --- a/guides/features/broadcast/arize.mdx +++ b/guides/features/broadcast/arize.mdx @@ -15,7 +15,9 @@ nofollow: false icon: puzzle-piece --- -[Arize AX](https://arize.com) is an evaluation and observability platform developed by Arize; it offers tools for agent tracing, evals, prompt optimization, and more. +Broadcast sends OpenRouter traces to [Arize AX](https://arize.com/docs/ax), Arize AI's enterprise product for observability and evaluation. AX is available as managed cloud or an enterprise self-hosted deployment. [Arize Phoenix](https://arize.com/docs/phoenix) is Arize AI's open-source product; the Broadcast configuration below is for AX. + +Use captured traces to review responses and investigate agent failures. Arize's guides cover [LLM evaluation](https://arize.com/resources/llm-evaluation/) and [agent evaluation](https://arize.com/guides/ai-agent-handbook/agent-evaluation/). ## Step 1: Get your Arize AX credentials