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, code, done. Set benchmark to "speed", title to "Typescript Utils", and done to true. Produce TypeScript utility code for grouping records by key, chunking arrays, and stable sorting with brief inline comments.
{
"metrics": {
"complete": {
"checks": [
{
"contains": [
"speed"
]
},
{
"contains": [
"done"
]
},
{
"contains": [
"typescript"
]
}
]
}
}
}temperature
0
max_tokens
620
timeout (s)
120
type
throughput
file
speed-typescript-utils.json