# Introducing Human-in-the-loop Evaluation for Agentic AI Observability

## The fastest path to quality labels
Spend less time labeling your data manually. Automatically classify and annotate text and images with AI that fits into your workflow.

## Rapidly pre-label text and images for large datasets
Kickstart your labeling process with off-the-shelf models, or use ground truth to generate predictions you can use as annotations for your entire project.

## Compare models for costs and quality
With Prompts versioning, you can test and evaluate which models are best suited for your project. Measure accuracy scores and inference costs or use LLM-as-a-judge to detect toxicity.

## Self-improving prompts
Prompt engineering can be tricky. Referencing your data and chosen LLM, Label Studio can enhance your prompts for more efficient, accurate results.

> 
>
> #### "With Prompts in Label Studio Enterprise, we've been able to bootstrap our labeling performance to near-human accuracy, transforming our data processing like never before."
>
> Dr. Tilo Sperling  
> Head of AI-Projects Business Applications

## Powered by

Prompts in Label Studio run on Adala, an open-source framework built specifically for data transformation. [Learn more](https://humansignal.com/blog/introducing-adala/).

- #### Verify results
- #### Prevent hallucinations
- #### Reduce inference costs

## Ready to put Prompts to work?
Try it on Starter Cloud, or talk to sales about adding Prompts to your plan.