Struct LlmLoadArgs 

pub struct LlmLoadArgs {

    pub url: String,
    pub model: String,
    pub concurrency: usize,
    pub duration: String,
    pub runtime_name: String,
    pub prompt_profile: Option<String>,
    pub prompt_file: Option<PathBuf>,
    pub warmup: String,
    pub output: Option<PathBuf>,
    pub stream: bool,
    pub rate: Option<f64>,
    pub rate_distribution: String,
    pub num_layers: Option<u32>,
    pub validate: String,
    pub fail_on_quality: Option<f64>,
    pub spike_threshold: f64,
    pub gpu_telemetry: bool,
    pub gpu_poll_interval: String,
    pub expected_clock_mhz: Option<u32>,
    pub skip_health_check: bool,
    pub dataset: Option<PathBuf>,
    pub max_tokens: Option<u32>,
    pub max_tokens_distribution: Option<String>,
}

Arguments for probador llm load

Fields

url: String

Base URL of the LLM API server

model: String

Model name to include in requests

concurrency: usize

Number of concurrent workers

duration: String

Test duration (e.g., 30s, 2m, 1h)

runtime_name: String

Runtime name for reporting

prompt_profile: Option<String>

Prompt profile: micro, short, medium, long

prompt_file: Option<PathBuf>

Path to YAML prompt file

warmup: String

Warmup duration before measurement (e.g., 5s, 10s). Default: no warmup.

output: Option<PathBuf>

Output file path for JSON results

stream: bool

Use SSE streaming for real per-token timing (TTFT, TPOT, ITL). Use –stream false to disable.

rate: Option<f64>

Target request rate (req/s). Omit for max throughput (closed-loop).

rate_distribution: String

Rate distribution: poisson (default) or constant. Only used with –rate.

num_layers: Option<u32>

Number of transformer layers in the model (e.g., 28 for Qwen 1.5B). Computes per-layer decode time for cross-runtime comparison.

validate: String

Inline correctness validation: none, basic, contains:X, pattern:X

fail_on_quality: Option<f64>

Exit non-zero if quality pass rate drops below this threshold (e.g., 0.95)

spike_threshold: f64

Multiplier of median ITL for spike detection (default: 5.0)

gpu_telemetry: bool

Enable GPU telemetry collection via nvidia-smi

gpu_poll_interval: String

GPU telemetry polling interval (e.g., 1s, 2s)

expected_clock_mhz: Option<u32>

Expected GPU clock speed in MHz for throttle detection (auto-detect if omitted)

skip_health_check: bool

Skip the pre-flight health check (not recommended)

dataset: Option<PathBuf>

Path to JSONL dataset file for workload-driven benchmarking

max_tokens: Option<u32>

Override max_tokens for all requests (e.g., –max-tokens 128)

max_tokens_distribution: Option<String>

Max tokens distribution: uniform:MIN,MAX or fixed:N. Creates heterogeneous traffic with staggered completion times. Example: –max-tokens-distribution uniform:16,128

Trait Implementations

impl Args for LlmLoadArgs

fn group_id() -> Option<Id>

Report the ArgGroup::id for this set of arguments

fn augment_args<'b>(__clap_app: Command) -> Command

Append to Command so it can instantiate Self via FromArgMatches::from_arg_matches_mut

fn augment_args_for_update<'b>(__clap_app: Command) -> Command

Append to Command so it can instantiate self via FromArgMatches::update_from_arg_matches_mut

impl CommandFactory for LlmLoadArgs

fn command<'b>() -> Command

Build a Command that can instantiate Self.

fn command_for_update<'b>() -> Command

Build a Command that can update self.

impl Debug for LlmLoadArgs

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

Formats the value using the given formatter.

impl FromArgMatches for LlmLoadArgs

fn from_arg_matches(__clap_arg_matches: &ArgMatches) -> Result<Self, Error>

Instantiate Self from ArgMatches, parsing the arguments as needed.

fn from_arg_matches_mut( __clap_arg_matches: &mut ArgMatches, ) -> Result<Self, Error>

Instantiate Self from ArgMatches, parsing the arguments as needed.

fn update_from_arg_matches( &mut self, __clap_arg_matches: &ArgMatches, ) -> Result<(), Error>

Assign values from ArgMatches to self.

fn update_from_arg_matches_mut( &mut self, __clap_arg_matches: &mut ArgMatches, ) -> Result<(), Error>

Assign values from ArgMatches to self.

impl Parser for LlmLoadArgs

fn parse() -> Self

Parse from std::env::args_os(), exit on error.

fn try_parse() -> Result<Self, Error>

Parse from std::env::args_os(), return Err on error.

fn parse_from<I, T>(itr: I) -> Self

where I: IntoIterator<Item = T>, T: Into<OsString> + Clone,

Parse from iterator, exit on error.

fn try_parse_from<I, T>(itr: I) -> Result<Self, Error>

where I: IntoIterator<Item = T>, T: Into<OsString> + Clone,

Parse from iterator, return Err on error.

fn update_from<I, T>(&mut self, itr: I)

where I: IntoIterator<Item = T>, T: Into<OsString> + Clone,

Update from iterator, exit on error.

fn try_update_from<I, T>(&mut self, itr: I) -> Result<(), Error>

where I: IntoIterator<Item = T>, T: Into<OsString> + Clone,

Update from iterator, return Err on error.