pub struct BenchmarkRun {
pub id: i64,
pub run_type: BenchmarkRunType,
pub status: BenchmarkRunStatus,
pub model_ids: Vec<i64>,
pub prompt_text: Option<String>,
pub system_prompt: Option<String>,
pub config_json: Option<String>,
pub applied_json: Option<String>,
pub error: Option<String>,
pub created_at: DateTime<Utc>,
pub completed_at: Option<DateTime<Utc>>,
}Expand description
Lightweight record grouping one or more model results under a single run.
Fields§
§id: i64Database ID of the run.
run_type: BenchmarkRunTypeWhether this is a compare, perf, or tune run.
status: BenchmarkRunStatusCurrent lifecycle state.
model_ids: Vec<i64>Ordered list of model IDs that were (or will be) benchmarked.
prompt_text: Option<String>Prompt text used for compare runs (absent for perf/tune runs).
system_prompt: Option<String>System prompt used for compare runs.
config_json: Option<String>Serialised run configuration (CompareConfig, PerfConfig, or
TuneConfig JSON).
applied_json: Option<String>The apply record written when this tune run’s winner became a model’s
Measured defaults (JSON-serialized tune::apply::ApplyRecord).
None on every run that was never applied — which is every run of
every other type, and most tune runs.
error: Option<String>Error message if the run failed.
created_at: DateTime<Utc>UTC timestamp when the run was created.
completed_at: Option<DateTime<Utc>>UTC timestamp when the run completed or failed.
Trait Implementations§
Source§impl Clone for BenchmarkRun
impl Clone for BenchmarkRun
Source§fn clone(&self) -> BenchmarkRun
fn clone(&self) -> BenchmarkRun
1.0.0 (const: unstable) · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
source. Read more