From 2bed66ccefa5caf90721d525f4ccac518113fe89 Mon Sep 17 00:00:00 2001 From: "alexa." Date: Thu, 27 Jun 2024 11:25:11 -0500 Subject: [PATCH] per sME review, remove call out in intro doc, update content in compatibility, add links to manual install --- .../compatibility-requirements-ai-monitoring.mdx | 4 ++-- src/content/docs/ai-monitoring/intro-to-ai-monitoring.mdx | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/src/content/docs/ai-monitoring/compatibility-requirements-ai-monitoring.mdx b/src/content/docs/ai-monitoring/compatibility-requirements-ai-monitoring.mdx index e929a14420c..e743b1c8114 100644 --- a/src/content/docs/ai-monitoring/compatibility-requirements-ai-monitoring.mdx +++ b/src/content/docs/ai-monitoring/compatibility-requirements-ai-monitoring.mdx @@ -86,9 +86,9 @@ AI monitoring is compatible with these agent versions and AI libraries: ## Monitoring at scale with NVIDIA NIM [#deploy-at-scale] -AI monitoring can integrate with NVIDIA NIM, allowing you to instrument your AI-powered apps and collect data about models even if those models aren't yet directly supported by AI monitoring. For example, if you've built a Python or Node.js AI app that uses llama3, mistralai, or one of NVIDIA's proprietary LLMs, you can still instrument with AI monitoring to view performance data about your apps. +AI monitoring can integrate with and collect data about any models supported by NVIDIA NIM. For example, if you've built a Python or Node.js AI app that uses llama3, mistralai, or one of NVIDIA's proprietary LLMs, you can instrument those apps with AI monitoring and view performance data about your apps. -No additional steps are needed to integrate with NVIDIA NIM: you can follow our manual procedures for installation, or you can [install directly through the New Relic platform](https://onenr.io/0VRVNLqavRa). +No additional steps are needed to integrate with NVIDIA NIM: you can follow our [manual procedures for installation](/install/ai-monitoring), or install [directly through the New Relic platform](https://onenr.io/0VRVNLqavRa). ## What's next? [#whats-next] diff --git a/src/content/docs/ai-monitoring/intro-to-ai-monitoring.mdx b/src/content/docs/ai-monitoring/intro-to-ai-monitoring.mdx index 13b33d1d859..88a62b324f7 100644 --- a/src/content/docs/ai-monitoring/intro-to-ai-monitoring.mdx +++ b/src/content/docs/ai-monitoring/intro-to-ai-monitoring.mdx @@ -43,8 +43,8 @@ Enabling AI monitoring allows the agent to recognize AI metadata associated with AI monitoring can help you answer critical questions about AI app performance: are your end users waiting too long for a response? Is there a recent spike in token usage? Are there patterns of negative user feedback around certain topics? With AI monitoring, you can see data specific to the AI-layer: * [Identify errors in specific prompt and response interactions](/docs/ai-monitoring/explore-ai-data/view-ai-responses) from the response table. If you're looking to make improvements to your AI models, [learn how to analyze your model data in New Relic](/docs/ai-monitoring/explore-ai-data/view-model-data). -* If you're using different models in different app environments, you can [compare the cost and performance of your apps before deploying](/docs/ai-monitoring/view-ai-data/#model-comparison). -* Do you deploy an AI app at scale with [NVIDIA NIM](https://www.nvidia.com/ai/)? AI monitoring can capture data about your [Python or Node.js AI-powered apps](https://onenr.io/0VRVNLqavRa). If NVIDIA NIM supports the model, then New Relic can integrate and capture the data. +* If you're using different models across app environments, you can [compare the cost and performance of your apps before deploying](/docs/ai-monitoring/view-ai-data/#model-comparison). +* Are you concerned about data compliance? [Learn how to create drop filters](/docs/ai-monitoring/drop-sensitive-data) to drop sensitive data before you send it to New Relic. ## Get started with AI monitoring [#get-started]