meta/llama-guard-4-12b

Capabilities

Reference ImagesSystem PromptTop-P

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`prompt`*	string	The prompt to send to the model. Do not use if using messages.	`—`	—
`system_prompt`*	string	System prompt to set the assistant's behavior	`—`	—
`frequency_penalty`	number	Frequency penalty parameter - positive values penalize the repetition of tokens.	`0`	min: -2, max: 2
`image_input`	array	List of images to send to the model		—
`max_completion_tokens`	integer	Maximum number of completion tokens to generate	`512`	min: 1, max: 1024
`presence_penalty`	number	Presence penalty parameter - positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.	`0`	min: -2, max: 2
`temperature`	number	Sampling temperature between 0 and 2	`1`	min: 0, max: 2
`top_p`	number	Nucleus sampling parameter - the model considers the results of the tokens with top_p probability mass. (0.1 means only the tokens comprising the top 10% probability mass are considered.)	`1`	min: 0, max: 1

promptrequiredstring

The prompt to send to the model. Do not use if using messages.

system_promptrequiredstring

System prompt to set the assistant's behavior

frequency_penaltynumber

Frequency penalty parameter - positive values penalize the repetition of tokens.

Default: 0min: -2, max: 2

image_inputarray

List of images to send to the model

Default:

max_completion_tokensinteger

Maximum number of completion tokens to generate

Default: 512min: 1, max: 1024

presence_penaltynumber

Presence penalty parameter - positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.

Default: 0min: -2, max: 2

temperaturenumber

Sampling temperature between 0 and 2

Default: 1min: 0, max: 2

top_pnumber

Nucleus sampling parameter - the model considers the results of the tokens with top_p probability mass. (0.1 means only the tokens comprising the top 10% probability mass are considered.)

Default: 1min: 0, max: 1

Version: b97828450982Updated: 7/25/202663.2K runs