Documentation Index
Fetch the complete documentation index at: https://arize-ax.mintlify.dev/docs/llms.txt
Use this file to discover all available pages before exploring further.
What’s New
April 25, 2024LangChain.js Auto Instrumentation
We’ve just released a new OpenInference package to support auto instrumentation of LangChain.js to unlock observability without the pain of manual instrumentation. Learn more here.Monitor Audit Log
The monitor audit log details all changes made to a monitor, including who made the change, the timestamp of the change, and specifics of what was altered. This helps teams track modifications and maintain oversight of monitor configurations and is available programmatically. Learn more here.Enhancements
April 25, 2024Update Actual and Tag Values
Arize supports the ability to update actual values and tag values. The process looks the same as sending a latent actual or tag.Prompt Playground Redesign
April 12, 2024 We’ve redesigned Prompt Playground to make it easier for users to compare prompts, models, and parameters. Additionally, Prompt Playground is now a top level page so it can be used as a blank slate testing ground!
Fileimporter Wildcard Support
Users can now use wildcard (**) in prefixes for file ingestion across GCS, S3 and Azure. Supports up to 3 layers of wildcards so users can define the ingestion pattern with more flexibility.- ** matches any sub-directories
- **abc** matches any sub-directories that contains “abc” anywhere
- abc** matches any sub-directories that starts with “abc”
- **abc matches any sub-directories that ends with “abc”

Light Mode (Beta)
Try out Light Mode, now available in beta. To turn light mode on, navigate to the bottom left corner of your screen, click the user icon -> select the sun 🔆.
Python SDK v7.14.1
- Supports export of spans from Arize platform
- Increases span field validation string length limits
- Allows spaces in eval names
📚 New Content
- Arize previews prompt variable monitoring at Google Cloud Next ‘24
- Research-driven tips for getting the “generation” part right with LLM RAG
- Chronos: learning the language of time series