Struct BenchResult 

pub struct BenchResult {

    pub name: String,
    pub difficulty: String,
    pub passed: bool,
    pub tool_calls: usize,
    pub input_tokens: u32,
    pub output_tokens: u32,
    pub cost_usd: Option<f64>,
    pub millis: u128,
    pub attempts: usize,
    pub turns: usize,
    pub trace: Vec<String>,
    pub answer_excerpt: String,
    pub answer: String,
    pub final_snapshot: String,
    pub error: Option<String>,
}

Outcome of running one BenchTask.

Fields

name: String

Task name.

difficulty: String

Difficulty label.

passed: bool

Whether the independent oracle judged the task complete.

tool_calls: usize

Tool calls the agent made.

input_tokens: u32

Input tokens consumed.

output_tokens: u32

Output tokens produced.

cost_usd: Option<f64>

Estimated cost in USD, if the provider reported it.

millis: u128

Wall-clock duration in milliseconds.

attempts: usize

Number of attempts made (1 = passed first try). Transient per-task failures (max-turns dither, a malformed model reply) are retried up to the suite’s max_attempts; this records how many it actually took. pass@k.

turns: usize

LLM turns observed via events — populated even when the task FAILS on a max-turns loop (unlike AgentOutput, lost on the error path).

trace: Vec<String>

Ordered tool-call sequence the agent issued (e.g. ["navigate_page", "take_snapshot", "click", ...]). Captured via events, so a FAILED task still shows what it looped on — the diagnostic the error path discarded.

answer_excerpt: String

First ~160 chars of the agent’s answer (for the trace).

answer: String

The agent’s FULL final answer (for an LLM judge / inspection).

final_snapshot: String

The independent post-run page snapshot used for grading (for an LLM judge).

error: Option<String>

Error, if the build or run failed before grading.

Trait Implementations

impl Clone for BenchResult

fn clone(&self) -> BenchResult

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for BenchResult

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

Formats the value using the given formatter.

impl Default for BenchResult

fn default() -> BenchResult

Returns the “default value” for a type.