Test fixture
Arithmetic, symbolic steps, and structured problem solving.
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 answer, evidence, constraints. A batch job runs every 6 hours starting at midnight UTC and takes up to 90 minutes to complete. A report is needed at 5:30am that must include data up to 4am. Will the report be ready at 5:30am?
{
"metrics": {
"accuracy": {
"checks": [
{
"contains": [
"no"
]
},
{
"contains": [
"not ready"
]
},
{
"contains": [
"7:30"
]
}
]
},
"evidence": {
"checks": [
{
"contains": [
"6am"
]
},
{
"contains": [
"90 minutes"
]
},
{
"contains": [
"midnight"
]
}
]
},
"constraint": {
"checks": [
{
"contains": [
"batch schedule"
]
},
{
"contains": [
"next run"
]
},
{
"contains": [
"finish time"
]
}
]
}
}
}temperature
0
max_tokens
500
timeout (s)
120
type
scored
file
reason-constraint-batch-window.json