Struct OllamaGenerateResponse 

pub struct OllamaGenerateResponse {
    pub text: String,
    pub context: Vec<i64>,
    pub prompt_eval_count: u64,
    pub eval_count: u64,
}

Full response from /api/generate, including KV cache context and token stats.

Used by OllamaClient::generate_with_full_response to support the two-step KV cache pattern (prime with document, then enrich each chunk cheaply).

Fields

text: String

The generated text

context: Vec<i64>

KV cache token state — pass back as OllamaGenerationParams::context on the next request to continue from this exact point without re-evaluating prior tokens.

prompt_eval_count: u64

Tokens actually evaluated in the prompt (vs reused from KV cache). With KV cache working: ~= chunk_tokens. Without: ~= full_prompt_tokens.

eval_count: u64

Tokens generated in the response.

Trait Implementations

impl Clone for OllamaGenerateResponse

fn clone(&self) -> OllamaGenerateResponse

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for OllamaGenerateResponse

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

Formats the value using the given formatter.