← Heatmap / Gemini 3.1 Flash Lite · Task 12
Gemini 3.1 Flash Lite · Tool use · Task 12 (Traps, missing month)
Mean -50.0%
| Field | Expected | Rollout #1 (-50%) | Rollout #2 (-50%) | Rollout #3 (-50%) |
|---|---|---|---|---|
| decision | clarify | calculate | calculate | calculate |
| currency | LCR | LCR | LCR | |
| gross pay | 5796.92 | 5796.92 | 5796.92 | |
| employee contribution | 245.00 | 245.00 | 245.00 | |
| allowed deduction | 0.00 | 0.00 | 0.00 | |
| taxable pay | 5551.92 | 5551.92 | 5551.92 | |
| income tax | 650.38 | 650.38 | 650.38 | |
| net pay | 4901.54 | 4901.54 | 4901.54 | |
| reason code | MISSING_PAYROLL_MONTH | |||
| clauses | R1 R7 | R1 R2 R3 R4 R5 R6 | R1 R2 R3 R4 R5 R6 | R1 R2 R3 R4 R5 R6 |
Prompt
Process this monthly payroll record under the supplied rulebook. Return the R6 or R7 JSON answer. Retrieve the case inputs with list_documents and read_document. Reference rows are available only through lookup_reference. Use calculate for exact decimal arithmetic.
Rollouts
Rollout #1 -50.0% answered trap · 4 tool calls · 5.6s · $0.0032
| Component | Score | Clauses |
|---|---|---|
| decision | 0.0% | R1 R7 |
| exact | 0.0% | R1 R7 |
| field accuracy | 0.0% | R1 R7 |
| citations | 0.0% | R1 R7 |
Assistant
list_documents()Tool
[
{"id": "20c5003e680b942a-payroll", "kind": "payroll"}
]
Assistant
read_document(document_id="20c5003e680b942a-payroll")Tool
{
"kind": "payroll",
"record": {"gross_pay": "5796.92", "allowed_deduction": "0.00"}
}
Assistant
lookup_reference(reference="first-half")Tool
{
"reference": "first-half",
"months": ["2026-01", "2026-06"],
"rate": "0.07",
"cap": "3500.00",
"deduction": "400.00",
"free": "1000.00",
"upper": "3600.00",
"middle_rate": "0.10",
"upper_rate": "0.20"
}
Assistant
calculate(program="employee_contribution = round(min(5796.92, 3500.00) * 0.07, 2)\nallowed_deduction = 0.00\ntaxable_pay = 5796.92 - employee_contribution - allowed_deduction\nincome_tax = round(max(0, min(taxable_pay - 1000.00, 3600.00 - 1000.00)) * 0.10 + max(0, taxable_pay - 3600.00) * 0.20, 2)\nnet_pay = 5796.92 - employee_contribution - allowed_deduction - income_tax")Tool
{
"results": {"employee_contribution": "245.00", "allowed_deduction": "0.00", "taxable_pay": "5551.92", "income_tax": "650.38", "net_pay": "4901.54"}
}
Assistant
{
"decision": "calculate",
"currency": "LCR",
"gross_pay": "5796.92",
"employee_contribution": "245.00",
"allowed_deduction": "0.00",
"taxable_pay": "5551.92",
"income_tax": "650.38",
"net_pay": "4901.54",
"clauses": ["R1", "R2", "R3", "R4", "R5", "R6"]
}
Rollout #2 -50.0% answered trap · 4 tool calls · 5.6s · $0.0033
| Component | Score | Clauses |
|---|---|---|
| decision | 0.0% | R1 R7 |
| exact | 0.0% | R1 R7 |
| field accuracy | 0.0% | R1 R7 |
| citations | 0.0% | R1 R7 |
Assistant
list_documents()Tool
[
{"id": "20c5003e680b942a-payroll", "kind": "payroll"}
]
Assistant
read_document(document_id="20c5003e680b942a-payroll")Tool
{
"kind": "payroll",
"record": {"gross_pay": "5796.92", "allowed_deduction": "0.00"}
}
Assistant
lookup_reference(reference="first-half")Tool
{
"reference": "first-half",
"months": ["2026-01", "2026-06"],
"rate": "0.07",
"cap": "3500.00",
"deduction": "400.00",
"free": "1000.00",
"upper": "3600.00",
"middle_rate": "0.10",
"upper_rate": "0.20"
}
Assistant
calculate(program="employee_contribution = round(min(5796.92, 3500.00) * 0.07, 2)\ntaxable_pay = 5796.92 - employee_contribution - 0.00\ntax_bracket1 = min(max(taxable_pay - 1000.00, 0), 2600.00)\ntax_bracket2 = max(taxable_pay - 3600.00, 0)\nincome_tax = round(tax_bracket1 * 0.10 + tax_bracket2 * 0.20, 2)\nnet_pay = 5796.92 - employee_contribution - 0.00 - income_tax")Tool
{
"results": {"employee_contribution": "245.00", "taxable_pay": "5551.92", "tax_bracket1": "2600.00", "tax_bracket2": "1951.92", "income_tax": "650.38", "net_pay": "4901.54"}
}
Assistant
{
"decision": "calculate",
"currency": "LCR",
"gross_pay": "5796.92",
"employee_contribution": "245.00",
"allowed_deduction": "0.00",
"taxable_pay": "5551.92",
"income_tax": "650.38",
"net_pay": "4901.54",
"clauses": ["R1", "R2", "R3", "R4", "R5", "R6"]
}
Rollout #3 -50.0% answered trap · 4 tool calls · 5.2s · $0.0032
| Component | Score | Clauses |
|---|---|---|
| decision | 0.0% | R1 R7 |
| exact | 0.0% | R1 R7 |
| field accuracy | 0.0% | R1 R7 |
| citations | 0.0% | R1 R7 |
Assistant
list_documents()Tool
[
{"id": "20c5003e680b942a-payroll", "kind": "payroll"}
]
Assistant
read_document(document_id="20c5003e680b942a-payroll")Tool
{
"kind": "payroll",
"record": {"gross_pay": "5796.92", "allowed_deduction": "0.00"}
}
Assistant
lookup_reference(reference="first-half")Tool
{
"reference": "first-half",
"months": ["2026-01", "2026-06"],
"rate": "0.07",
"cap": "3500.00",
"deduction": "400.00",
"free": "1000.00",
"upper": "3600.00",
"middle_rate": "0.10",
"upper_rate": "0.20"
}
Assistant
calculate(program="employee_contribution = round(min(5796.92, 3500.00) * 0.07, 2)\nallowed_deduction = 0.00\ntaxable_pay = 5796.92 - employee_contribution - allowed_deduction\nincome_tax = round(max(0, min(taxable_pay, 3600.00) - 1000.00) * 0.10 + max(0, taxable_pay - 3600.00) * 0.20, 2)\nnet_pay = 5796.92 - employee_contribution - allowed_deduction - income_tax")Tool
{
"results": {"employee_contribution": "245.00", "allowed_deduction": "0.00", "taxable_pay": "5551.92", "income_tax": "650.38", "net_pay": "4901.54"}
}
Assistant
{
"decision": "calculate",
"currency": "LCR",
"gross_pay": "5796.92",
"employee_contribution": "245.00",
"allowed_deduction": "0.00",
"taxable_pay": "5551.92",
"income_tax": "650.38",
"net_pay": "4901.54",
"clauses": ["R1", "R2", "R3", "R4", "R5", "R6"]
}