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 request asks for zero downtime, no dual writes, no read fallback, and immediate rollback for a breaking schema change. Explain the conflict.
{
"metrics": {
"accuracy": {
"checks": [
{
"contains": [
"zero downtime"
]
},
{
"contains": [
"no dual writes"
]
},
{
"contains": [
"breaking schema"
]
}
]
},
"evidence": {
"checks": [
{
"contains": [
"no read fallback"
]
},
{
"contains": [
"rollback"
]
},
{
"contains": [
"conflict"
]
}
]
},
"constraint": {
"checks": [
{
"contains": [
"uncertainty"
]
},
{
"contains": [
"impossible"
]
},
{
"contains": [
"tradeoff"
]
}
]
}
}
}temperature
0
max_tokens
500
timeout (s)
120
type
scored
file
reasoning-uad-impossible-migration.json