Struct GenerationConfig 

pub struct GenerationConfig {

    pub temperature: Option<f32>,
    pub top_p: Option<f32>,
    pub top_k: Option<i32>,
    pub max_output_tokens: Option<i32>,
    pub candidate_count: Option<i32>,
    pub stop_sequences: Option<Vec<String>>,
    pub response_mime_type: Option<String>,
    pub response_schema: Option<Value>,
    pub response_modalities: Option<Vec<String>>,
    pub image_config: Option<ImageConfig>,
    pub speech_config: Option<SpeechConfig>,
    pub thinking_config: Option<ThinkingConfig>,
    pub media_resolution: Option<MediaResolutionLevel>,
}

Configuration for generation

Fields

temperature: Option<f32>

The temperature for the model (0.0 to 1.0)

Controls the randomness of the output. Higher values (e.g., 0.9) make output more random, lower values (e.g., 0.1) make output more deterministic.

top_p: Option<f32>

The top-p value for the model (0.0 to 1.0)

For each token generation step, the model considers the top_p percentage of probability mass for potential token choices. Lower values are more selective, higher values allow more variety.

top_k: Option<i32>

The top-k value for the model

For each token generation step, the model considers the top_k most likely tokens. Lower values are more selective, higher values allow more variety.

max_output_tokens: Option<i32>

The maximum number of tokens to generate

Limits the length of the generated content. One token is roughly 4 characters.

candidate_count: Option<i32>

The candidate count

Number of alternative responses to generate.

stop_sequences: Option<Vec<String>>

Whether to stop on specific sequences

The model will stop generating content when it encounters any of these sequences.

response_mime_type: Option<String>

The response mime type

Specifies the format of the model’s response.

response_schema: Option<Value>

The response schema

Specifies the JSON schema for structured responses.

response_modalities: Option<Vec<String>>

Response modalities (for TTS and other multimodal outputs)

image_config: Option<ImageConfig>

Optional. Config for image generation. An error will be returned if this field is set for models that don’t support these config options.

speech_config: Option<SpeechConfig>

Speech configuration for text-to-speech generation

thinking_config: Option<ThinkingConfig>

The thinking configuration

Configuration for the model’s thinking process (Gemini 2.5 and Gemini 3 series).

media_resolution: Option<MediaResolutionLevel>

Global media resolution for all images and PDFs. Controls the resolution used for inline image and PDF data, affecting token usage. Can be overridden per-part using the Part::InlineData media_resolution field.

Trait Implementations

impl Clone for GenerationConfig

fn clone(&self) -> GenerationConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GenerationConfig

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

Formats the value using the given formatter.

impl Default for GenerationConfig

fn default() -> GenerationConfig

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for GenerationConfig

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for GenerationConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.