What is a Test Case?

A test case in Galtea is a challenge designed to evaluate the performance of a product. It represents a specific input and context sent to an AI model alongside an expected output that can be used to assess the product’s capabilities.

Using Test Cases in Evaluations

Test cases are used with evaluation tasks to assess the performance of specific versions of your product.

Create an Evaluation

Learn how to use tests in evaluations

SDK Integration

Test Case Service SDK

SDK methods for managing test cases

Test Case Properties

When creating a test case in Galtea, you’ll need to provide the following information:

test_id
string
required

The ID of the test you want to create the test case for.

input
string
required

The input data used for inference on the LLM product’s version. Example: “How much does the iPhone 16 cost?”

expected_output
string

The expected output for the evaluation task, extracted from the test file. Example: “The iPhone 16 costs $999.”

context
string

Additional context provided to the product’s version upon inference alongside the input. Example: “[previous messages of the conversation]”

It is test-case-specific and provides context to the model. Do not mistake for the system prompt/few-shot examples; those should be defined in the product’s version unless they change on an interaction basis.

tag
string

The tag for the test case. This is a label that can help you categorize or identify the test case.

source
string

The original source text used to create the test case. This field is particularly useful for tracing the origin of the information used in the test case.

Typically, this is only filled if the test case is created by Galtea’s Test Case Generator, but you can also use it to provide the source of the test case if you created it manually.