05.10.2024

New Releases, Enhancements, + Changes

What's New

Sessions

A session is a grouping of traces based on a session ID attribute. When building or debugging a chatbot application, being able to see groups of messages or traces belonging to a series of interactions between a human and the AI can be particularly helpful. By adding session.id and user.id as attributes to spans, you can:

• Find exactly where a conversation "breaks" or goes off the rails. This can help identify if a user becomes progressively more frustrated or if a chatbot is not helpful.

• Find groups of traces where your application is not performing well. Adding session.id and/or user.id from an application enables back-and-forth interactions to be grouped and filtered further.

• Construct custom metrics based on evals using session.id or user.id to find best/worst performing sessions and users.

Learn more ->

OpenAI GPT-4o Support

Test out the newest model from OpenAI, GPT-4o, now available in Prompt Playground in Arize. With vision / image support, prompt template logging, and tracing you can monitor and iterate on your LLM applications using GPT-4.

Enhancements

Join Evals on Existing Traces

• Send in latent evals through our Python SDK. They will be joined daily on spans with matching span_ids. The functionality is very similar to that of latent actuals or tags, but with Evals.

• Send in your evaluations through the new log_evaluations method in the new Python SDK v7.16.1.

Learn more ->

Programmatically Create Dashboards

Replicate a particular set of widgets across various dashboards, clone an entire dashboard, or integrate dashboard creation into your automated onboarding workflows with more ease. Learn more ->

Homepage Refresh

Designed to streamline your workflow, so you can hop right into tracking and troubleshooting your projects.

Python SDK v.7.17.0

• Add session.id and user.id to spans batch logging

• Send Arize schema as part of the request body

Learn about Python SDK fixes and improvements here.

📚 New Content

The latest paper readings, ebooks, self-guided learning modules, and technical posts:

• Four tips on how to read AI research papers effectively

• ReAct: Synergizing Reasoning and Acting in Language Models

• LLM Time Series: Research on Performance of GPT-4 and Claude 3 At Time Series Analysis

• Citi: Good Things Happen Podcast on AI