Struct AiEvalResultRecord 

pub struct AiEvalResultRecord {
    pub ts_ms: i64,
    pub eval_name: String,
    pub model: String,
    pub variant: String,
    pub prompt_version: String,
    pub case_index: usize,
    pub input: String,
    pub expected: String,
    pub actual: String,
    pub passed: bool,
    pub scorer: String,
    pub latency_ms: f64,
}

One AI eval result record (DW-086): the per-case record written to the ai_eval_results table when an eval is run via the admin API. A PLAIN DTO — the admin endpoint constructs it from the eval runner’s output.

Fields

ts_ms: i64

Wall-clock ms since the Unix epoch.

eval_name: String

The eval name (from config).

model: String

The model alias the eval ran against.

variant: String

The variant name (when running an A/B test’s variants), or empty.

prompt_version: String

The prompt version reference (prompt_name/version_name), or empty.

case_index: usize

The case index in the golden set.

input: String

The input prompt.

expected: String

The expected output.

actual: String

The actual output from the provider.

passed: bool

Whether the case passed (scorer matched).

scorer: String

The scorer name (exact_match, contains, regex).

latency_ms: f64

The latency of the provider call in milliseconds.

Trait Implementations

impl Clone for AiEvalResultRecord

fn clone(&self) -> AiEvalResultRecord

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AiEvalResultRecord

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

Formats the value using the given formatter.