Agenta prompt playground now supports Jinja2 prompt templates. Create dynamic LLM prompts with conditional logic. Prompt management with Jinja2 templating support.

Today we're open-sourcing most of our product. Here's what we learned after three failed attempts.

We're open-sourcing all functional features of Agenta under the MIT license.

We're launching the evaluation SDK today. The evaluation SDK allows you to evaluate complex agents and LLM workflows using built-in or custom evaluators.

Today we're launching online evaluation. With online evaluation each request is automatically evaluated. This allows you to monitor things in production.

Launch Week Day 1: Redesigned evaluation dashboard with side-by-side comparison, detailed debugging, and customizable LLM-as-a-judge evaluators.