Changelog
Back to All

10/23/24 Updates

5 months ago by Jeremy Silva

💡

Recent updates: New Eval Template Library & AI Assistant, New Usage Dashboards & Spend Management, Model Updates Including Claude 3.5 Sonnet & Llama 3.2, Plus Major UI Improvements, Bug Fixes, & Significant Performance Improvements

  • Eval suite upgraded with template library and an AI assistant (More on the blog)
  • Monitor team spend and usage across providers and environments (More on the blog)
  • Run auto evals on any individual session straight from the observability tab
  • Model updates: Claude 3.5 Sonnet on Anthropic, plus Llama 3.2 on Bedrock
  • Major UI improvements, plus bug fixes and a significant boost in performance

Read more here

Write Better Evals Faster with Templates & AI Assistance

We’ve heard from customers that they often know what they want to evaluate but struggle to implement these evaluations effectively so we added AI assistance throughout the eval creation process. Our new AI assistance feature for eval templates will help you create customized evaluations that leverage all the latest best practices.

You can read more about these features on our blog, plus learn how we’re using AI to help your team move faster with evals.

Dig Into Team-Wide Usage & Spend Management Dashboards

We've given teams the ability to track their usage across the Freeplay app. You'll now see a new Usage tab in the app where you can find a breakdown of all your costs and usage metrics by prompt. Read about these new features on our blog.

Faster & Smoother Product Experience Thanks to UI & Performance Improvements

We've made big investments in the core UI over the last month, making for a more streamlined user experience. We’ve also invested in major performance improvements resulting in up to 20x speed improvements for our largest scale customers.

Additional Updates

  • You can now run auto evals on any session straight from the observability tab by clicking the "Run auto evals" button from any prompt screen.
  • We’ve added support for a handful of new models including Claude 3.5 Sonnet on Anthropic, plus Llama 3.2 on Bedrock.