> For the complete documentation index, see [llms.txt](https://docs.datumo.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.datumo.com/documentation/documentation-en/readme/quickstart.md).
# Quickstart
{% tabs %}
{% tab title="Evaluation Quick Start" %}
{% hint style="info" %}
Evaluation Quick Start? Head to the [Basics](/documentation/documentation-en/evaluation/editor.md) section to learn more.
{% endhint %}
### Create Workspace & Application
You must create a Workspace and register the Application to be evaluated.
Since the Application is the basic unit of evaluation and monitoring, you must first define which AI service to evaluate so that you can later connect settings such as Criteria, Benchmark Set, Metric, and Run Evaluation.
### **Connect Judge Model**
Register the LLM Judge model to use for evaluation.\
The Judge Model is used to produce evaluation results in LLM-as-a-Judge-based Metrics.
In the pre-trial environment, a default Judge Model is pre-registered, so you can use it without additional setup.
If you use a separate model, register the Judge Model by entering the required information such as the API URL, API Key, and model name in the Chat Completion API format.
### **Create Criteria & Add Template Task**
After creating the Criteria that will serve as the evaluation standard, add a Template Task.\
Using a Template Task configures the default settings for the Metric and Benchmark Set together, so you can set up the evaluation environment faster than creating a Custom Task directly.
### **Run Eval**
Proceed with the evaluation.
### **Check Results**
You can check the Score in Eval Result.
{% endtab %}
{% tab title="Red Teaming Quick Start" %}
#### Select Benchmark Set
Select the Benchmark Set to use for Task evaluation. You can review the built-in Benchmark Sets and choose data that fits your evaluation purpose.
#### Run Auto Redteaming
After setting the Attack Method and target model to use for evaluation, run the red teaming evaluation.
#### Check Results
After running the evaluation, you can check the attack success rate and failure rate in the Dashboard, and analyze the model's safety vulnerabilities.
{% endtab %}
{% tab title="Data Generation Quick Start" %}
You can use AI to generate Benchmark Set data or modify existing data.
#### Select Generation Template
#### Enter Generation Conditions
Enter the columns to generate and the information needed for data generation. \
The entered conditions are used as the criteria for AI to generate Benchmark Set data.
If needed, you can add a Draft to make the instructions more specific.
#### Generate Data
Generate data based on the entered conditions. \
After reviewing the generated results, you can save them as Benchmark Set data if you are satisfied.
{% endtab %}
{% endtabs %}