Test fixture
Throughput and TTFT-focused generation tasks.
The model receives the prompt (and optional system message). The run uses scorer rubric_json_metrics with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.
Return JSON only with keys benchmark, title, items, done. Set benchmark to "speed", title to "Log Summary", and done to true. Summarize these log themes into 10 concise items: retries, cache misses, slow queries, queue depth, timeouts.
{
"metrics": {
"complete": {
"checks": [
{
"contains": [
"speed"
]
},
{
"contains": [
"done"
]
},
{
"contains": [
"log"
]
}
]
}
}
}temperature
0
max_tokens
360
timeout (s)
120
type
throughput
file
speed-log-summary.json