Struct Prompt 

pub struct Prompt {
    pub instruction: String,
    pub input: String,
    pub max_output_tokens: u32,
    pub temperature: f32,
}

What the application asks a model to do.

Deliberately not a chat transcript: most product features are one instruction over one piece of content, and modelling them as conversations invites state nobody needs.

Fields

instruction: String

What the model should do.

input: String

What it should do it to.

max_output_tokens: u32

Upper bound on the answer. A product that does not cap this is one bad prompt away from a surprising bill.

temperature: f32

0.0 for extraction and classification, higher for drafting.

Implementations

impl Prompt

pub fn new(instruction: impl Into<String>, input: impl Into<String>) -> Self

A deterministic prompt: temperature zero, short answer.

The right default for classification and extraction, which is most of what a desktop application actually needs.

pub fn with_max_output_tokens(self, max_output_tokens: u32) -> Self

pub fn with_temperature(self, temperature: f32) -> Self

Allow variation. Use for drafting, never for extraction.

Trait Implementations

impl Clone for Prompt

fn clone(&self) -> Prompt

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Prompt

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for Prompt

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for Prompt

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.