Skip to main content
JSON Schema provides several keywords for validating numeric instances (both number and integer types).

​
multipleOf

Validates that a number is a multiple of a specified value.

​
Syntax

The value must be a number strictly greater than 0. 
{
  "multipleOf": 5
}

​
Validation Rules

A numeric instance is valid only if division by the multipleOf value results in an integer.

​
Examples

​
Integer Multiples

{
  "type": "number",
  "multipleOf": 5
}
Valid instances: 
0
5
10
-15
Invalid instances: 
3
7
12

​
Decimal Multiples

{
  "type": "number",
  "multipleOf": 0.5
}
Valid instances: 
0.5
1.0
2.5
-1.5
Invalid instances: 
0.3
1.7
2.25

​
Even Numbers

{
  "type": "integer",
  "multipleOf": 2
}
Valid instances: 
0
2
-4
100
Invalid instances: 
1
3
-5

​
minimum

Specifies an inclusive lower bound for a numeric value.

​
Syntax

The value must be a number. 
{
  "minimum": 0
}

​
Validation Rules

If the instance is a number, it validates only if the value is greater than or equal to minimum.

​
Examples

{
  "type": "number",
  "minimum": 0
}
Valid instances: 
0
1
100.5
Invalid instances: 
-1
-0.01
-100

​
maximum

Specifies an inclusive upper bound for a numeric value.

​
Syntax

The value must be a number. 
{
  "maximum": 100
}

​
Validation Rules

If the instance is a number, it validates only if the value is less than or equal to maximum.

​
Examples

{
  "type": "number",
  "maximum": 100
}
Valid instances: 
0
50
100
Invalid instances: 
100.1
101
200

​
exclusiveMinimum

Specifies an exclusive lower bound for a numeric value.

​
Syntax

The value must be a number. 
{
  "exclusiveMinimum": 0
}

​
Validation Rules

If the instance is a number, it is valid only if the value is strictly greater than (not equal to) exclusiveMinimum.

​
Examples

{
  "type": "number",
  "exclusiveMinimum": 0
}
Valid instances: 
0.01
1
100
Invalid instances: 
0
-1
-0.01

​
exclusiveMaximum

Specifies an exclusive upper bound for a numeric value.

​
Syntax

The value must be a number. 
{
  "exclusiveMaximum": 100
}

​
Validation Rules

If the instance is a number, it is valid only if the value is strictly less than (not equal to) exclusiveMaximum.

​
Examples

{
  "type": "number",
  "exclusiveMaximum": 100
}
Valid instances: 
99.99
50
0
Invalid instances: 
100
100.01
200

​
Combining Numeric Keywords

Multiple numeric validation keywords can be used together to define precise ranges and constraints.

​
Age Range

{
  "type": "integer",
  "minimum": 0,
  "maximum": 120
}
Valid instances: 
0
25
120

​
Percentage

{
  "type": "number",
  "minimum": 0,
  "maximum": 100,
  "multipleOf": 0.01
}
This validates percentages with up to 2 decimal places: 
0
50.5
99.99
100

​
Temperature Range (exclusive bounds)

{
  "type": "number",
  "exclusiveMinimum": -273.15,
  "maximum": 1000
}
This validates temperatures above absolute zero up to 1000: 
-273.14
0
100.5
1000
Invalid: 
-273.15  // Equal to exclusiveMinimum
1000.01  // Exceeds maximum

​
Price Validation

{
  "type": "number",
  "minimum": 0,
  "exclusiveMinimum": 0,
  "multipleOf": 0.01
}
This validates positive prices with cent precision: 
0.01
9.99
100.00
Invalid: 
0       // Not greater than exclusiveMinimum
-1      // Negative price
10.999  // Too many decimal places

​
Score Range

{
  "type": "integer",
  "minimum": 0,
  "maximum": 100,
  "multipleOf": 5
}
Valid instances (scores in increments of 5): 
0
5
50
100
Invalid instances: 
3
47
105

​
Positive Integer

{
  "type": "integer",
  "minimum": 1
}
Valid instances: 
1
10
1000
Invalid instances: 
0
-1
3.14

​
Open Range

{
  "type": "number",
  "exclusiveMinimum": 0,
  "exclusiveMaximum": 1
}
This validates numbers in the open interval (0, 1): 
0.1
0.5
0.999
Invalid: 
0
1

​
Precision Considerations

JSON Schema allows numbers with arbitrary precision. However, be aware that:
  • Floating-point arithmetic may introduce rounding errors
  • Programming language limitations may affect precision
  • When using multipleOf with decimal values, consider potential floating-point representation issues
For financial calculations, consider using integer types with an implied decimal point (e.g., storing cents as integers).

Build docs developers (and LLMs) love

Get started for freeTalk to us