Enum PerplexityModels 

pub enum PerplexityModels {
    SonarPro,
    Sonar,
    SonarReasoning,
    Llama3_1SonarSmall,
    Llama3_1SonarLarge,
    Llama3_1SonarHuge,
}

Variants

SonarPro

Sonar

SonarReasoning

Llama3_1SonarSmall

👎Deprecated since 0.12.0: Llama3_1SonarSmall is deprecated starting February 22, 2025, please use Sonar or SonarPro instead.

Llama3_1SonarLarge

👎Deprecated since 0.12.0: Llama3_1SonarLarge is deprecated starting February 22, 2025, please use Sonar or SonarPro instead.

Llama3_1SonarHuge

👎Deprecated since 0.12.0: Llama3_1SonarHuge is deprecated starting February 22, 2025, please use Sonar or SonarPro instead.

Trait Implementations

impl Clone for PerplexityModels

fn clone(&self) -> PerplexityModels

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PerplexityModels

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for PerplexityModels

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl LLMModel for PerplexityModels

fn call_api<'life0, 'life1, 'life2, 'life3, 'async_trait>( &'life0 self, api_key: &'life1 str, _version: Option<String>, body: &'life2 Value, debug: bool, _tools: Option<&'life3 [LLMTools]>, ) -> Pin<Box<dyn Future<Output = Result<String>> + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, 'life3: 'async_trait,

This function leverages Perplexity API to perform any query as per the provided body.

It returns a String the Response object that needs to be parsed based on the self.model.

fn sanitize_json_response(&self, json_response: &str) -> String

This function sanitizes the text response from Perplexity models to clean up common formatting issues. Currently the function checks:

• json{} wrapper around response
• wrapper (for SonarReasoning model only)

fn as_str(&self) -> &str

Converts each item in the model enum into its string representation

fn try_from_str(name: &str) -> Option<Self>

Returns an instance of the enum based on the provided string representation of name

fn default_max_tokens(&self) -> usize

Returns max supported number of tokens for each of the variants of the enum

fn get_endpoint(&self) -> String

Returns the url of the endpoint that should be called for each variant of the LLM Model enum

fn get_body( &self, instructions: &str, json_schema: &Value, function_call: bool, _max_tokens: &usize, temperature: &f32, _tools: Option<&[LLMTools]>, _thinking_level: Option<&ThinkingLevel>, ) -> Value

Constructs the body that should be attached to the API call for each of the LLM Models

fn get_data(&self, response_text: &str, _function_call: bool) -> Result<String>

Based on the model type extracts the data portion of the API response

fn get_rate_limit(&self) -> RateLimit

Returns the rate limit accepted by the API depending on the used model If not explicitly defined it will assume 1B tokens or 100k transactions a minute

fn get_normalized_temperature(&self, relative_temp: u32) -> f32

Returns the normalized temperature for the model

fn get_version_endpoint(&self, _version: Option<String>) -> String

Returns the url of the endpoint that should be called for each variant of the LLM Model enum It allows to specify which version of the endpoint to use

fn get_base_instructions(&self, _function_call: Option<bool>) -> String

Provides a list of base instructions that should be added to each prompt when using each of the models

fn function_call_default(&self) -> bool

Returns recommendation if function calling should be used for the specified model

fn get_version_body( &self, instructions: &str, json_schema: &Value, function_call: bool, max_tokens: &usize, temperature: &f32, _version: Option<String>, tools: Option<&[LLMTools]>, thinking_level: Option<&ThinkingLevel>, ) -> Value

An API-version-specific implementation of the body constructor

fn get_version_data( &self, response_text: &str, function_call: bool, _version: Option<String>, ) -> Result<String>

An API-version-specific implementation of the data extractor

fn get_max_requests(&self) -> usize

Based on the RateLimit for the model calculates how many requests can be send to the API

fn get_default_temperature(&self) -> f32

Returns the default temperature to be used by the model

impl PartialEq for PerplexityModels

fn eq(&self, other: &PerplexityModels) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for PerplexityModels

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Eq for PerplexityModels

impl StructuralPartialEq for PerplexityModels