Added Groq Model, added several site wide eval templates

Added Groq Model, added several site wide eval templates

Date
May 7, 2025
  • Added the ability for customers to use Grok model with the Llama 3.3 70B Versatile version with Empromptu AI
  • We also added several site wide standard eval templates based on customer feedback
    • Test and ensure the AI response output is in a "Professional Tone" when using AI. The tone aligns with the context—whether formal business communication, academic writing, or technical documentation.
    • Test for correct sequence of information. Ensuring that AI outputs are in the correct order for math equations etc
    • Test for if the AI output gives incomplete information. The extracted AI output contains all the same information as the original despite being in a different form.