A simple endpoint to check if the Prediction Guard API is healthy.

Check API Health

Generate chat completions based on a conversation history.

example

The chat model to use for generating completions.

The conversation history with roles (user, assistant) and messages.

The maximum number of tokens in the generated completion.

The temperature parameter for controlling randomness in completions.

Generate Chat Completions

Retrieve text completions based on the provided input.

Generate Completions

Check the factuality of a given text compared to a reference.

Check Factuality

Translate text from one language to another.

Translate Text

Replace personal information such as names, SSNs, and emails in a given text.

The text containing personal information to be replaced.

Specify whether to replace personal information or not.

The method to use for replacing personal information.

Replace Personal Information

Detect potential prompt injection attacks in a given prompt.

The prompt text to detect injection attacks.

Specify whether to detect potential injection attacks.

Detect Injection

Check Toxicity

CompletionsMessages

The role of the sender (user or assistant).

InlineResponse200

Timestamp of when the chat completion was created.

The chat model used for generating completions.

InlineResponse2001

Timestamp of when the factuality check was created.

InlineResponse2001Checks

The factuality score for a specific statement.

The index of the statement in the provided text.

The status of the factuality check for a specific statement.

InlineResponse2002

InlineResponse2002Response

InlineResponse2003

Timestamp of when the PII check was created.

InlineResponse2003Checks

The text with replaced personal information.

InlineResponse2004

Timestamp of when the injection check was created.

InlineResponse2004Checks

The probability of a potential injection attack.

InlineResponse2005

Timestamp of when the toxicity check was created.

InlineResponse2005Checks

The toxicity score for the provided text.

The status of the toxicity check for the provided text.

InlineResponse200Choices

InlineResponse200Message

PostCompletionsResponse

Timestamp of when the completion was created.

PostCompletionsResponseChoicesItem

The model used for generating the completion.

1	curl https://api.predictionguard.com \
2	-H "x-api-key: <apiKey>"