{
  "op": "ai.onnx.Constant",
  "cases": [
    {
      "name": "ort_multivalue_constant_floats",
      "provenance": {
        "source": "onnxruntime/test/providers/cpu/tensor/constant_test.cc",
        "test": "ConstantOpTest.MultiValueConstant_floats"
      },
      "attrs": { "value_floats": [0, 1, 2, 3] },
      "inputs": {},
      "outputs": { "output": { "dtype": "float32", "shape": [4], "tolerance": 0 } }
    },
    {
      "name": "f32_subnormal_literal_values",
      "provenance": {
        "source": "onnxruntime/test/providers/cpu/tensor/constant_test.cc",
        "test": "ConstantOpTest.MultiValueConstant_floats",
        "notes": "A Constant node with explicit finite subnormal float payloads should materialize those bit patterns without arithmetic."
      },
      "attrs": { "value": { "dtype": "float32", "shape": [4], "values": [1e-40, -1e-40, 4e-39, -4e-39] } },
      "inputs": {},
      "outputs": {
        "output": {
          "dtype": "float32",
          "shape": [4],
          "tolerance": 0,
          "data": { "kind": "values", "values": [1e-40, -1e-40, 4e-39, -4e-39] }
        }
      }
    },
    {
      "name": "ort_multivalue_constant_ints_int32",
      "provenance": {
        "source": "onnxruntime/test/providers/cpu/tensor/constant_test.cc",
        "test": "ConstantOpTest.MultiValueConstant_ints",
        "notes": "Native int32 projection of ORT's int64 value_ints shape/value coverage."
      },
      "attrs": { "value": { "dtype": "int32", "shape": [4], "values": [0, 1, 2, 3] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "int32", "shape": [4], "tolerance": 0 } }
    },
    {
      "name": "literal_i32_mixed_exact_values",
      "attrs": { "value": { "dtype": "int32", "shape": [2, 2], "values": [-123456789, -1, 0, 16777217] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "int32", "shape": [2, 2], "tolerance": 0 } }
    },
    {
      "name": "literal_u32_exact_values",
      "attrs": { "value": { "dtype": "uint32", "shape": [2, 2], "values": [0, 1, 16777217, 4000000000] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "uint32", "shape": [2, 2], "tolerance": 0 } }
    },
    {
      "name": "literal_int8_edge_values",
      "attrs": { "value": { "dtype": "int8", "shape": [2, 2], "values": [-128, -1, 0, 127] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "int8", "shape": [2, 2], "tolerance": 0 } }
    },
    {
      "name": "literal_uint8_edge_values",
      "attrs": { "value": { "dtype": "uint8", "shape": [2, 2], "values": [0, 1, 254, 255] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "uint8", "shape": [2, 2], "tolerance": 0 } }
    },
    {
      "name": "literal_int16_edge_values",
      "provenance": {
        "source": "onnx/onnx/docs/Operators.md#Constant-24",
        "notes": "Covers the complete signed int16 range through Constant's literal-value path."
      },
      "attrs": { "value": { "dtype": "int16", "shape": [2, 2], "values": [-32768, -1, 0, 32767] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "int16", "shape": [2, 2], "tolerance": 0 } }
    },
    {
      "name": "literal_bool_values",
      "provenance": {
        "source": "onnx/onnx/docs/Operators.md#Constant-24",
        "notes": "ONNX bool tensor payloads use the repository's canonical widened 0/1 representation."
      },
      "attrs": { "value": { "dtype": "bool", "shape": [2, 2], "values": [0, 1, 1, 0] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "bool", "shape": [2, 2], "tolerance": 0 } }
    },
    {
      "name": "literal_f16_values",
      "attrs": { "value": { "dtype": "float16", "shape": [4], "values": [-2, -0.5, 0.25, 3.5] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "float16", "shape": [4], "tolerance": 0.001 } }
    },
    {
      "name": "fill_f32_scalar_attr",
      "attrs": { "value_float": -2.5 },
      "inputs": {},
      "outputs": { "output": { "dtype": "float32", "shape": [] } }
    },
    {
      "name": "fill_uint32_scalar_attr",
      "attrs": { "value": { "dtype": "uint32", "shape": [4], "values": [7, 7, 7, 7] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "uint32", "shape": [4] } }
    },
    {
      "name": "fill_f16_scalar_attr",
      "attrs": { "value": { "dtype": "float16", "shape": [2, 2], "values": [1.25, 1.25, 1.25, 1.25] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "float16", "shape": [2, 2] } },
      "tolerance": 0.001
    },
    {
      "name": "fill_int8_negative_scalar_attr",
      "attrs": { "value": { "dtype": "int8", "shape": [2, 3], "values": [-128, -128, -128, -128, -128, -128] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "int8", "shape": [2, 3], "tolerance": 0 } }
    },
    {
      "name": "fill_uint8_high_scalar_attr",
      "attrs": { "value": { "dtype": "uint8", "shape": [2, 3], "values": [250, 250, 250, 250, 250, 250] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "uint8", "shape": [2, 3], "tolerance": 0 } }
    },
    {
      "name": "fill_int16_min_scalar_attr",
      "provenance": {
        "source": "onnx/onnx/docs/Operators.md#Constant-24",
        "notes": "Exercises the scalar-fill path at the signed int16 minimum."
      },
      "attrs": {
        "value": { "dtype": "int16", "shape": [2, 3], "values": [-32768, -32768, -32768, -32768, -32768, -32768] }
      },
      "inputs": {},
      "outputs": { "output": { "dtype": "int16", "shape": [2, 3], "tolerance": 0 } }
    },
    {
      "name": "fill_bool_true_scalar_attr",
      "provenance": {
        "source": "onnxruntime/test/framework/function_test.cc",
        "test": "Constant <value = bool {1}>",
        "notes": "Exercises the scalar-fill path for a true ONNX bool constant."
      },
      "attrs": { "value": { "dtype": "bool", "shape": [2, 3], "values": [1, 1, 1, 1, 1, 1] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "bool", "shape": [2, 3], "tolerance": 0 } }
    },
    {
      "name": "fill_i32_negative_exact",
      "attrs": {
        "value": { "dtype": "int32", "shape": [2, 2], "values": [-123456789, -123456789, -123456789, -123456789] }
      },
      "inputs": {},
      "outputs": { "output": { "dtype": "int32", "shape": [2, 2], "tolerance": 0 } }
    },
    {
      "name": "fill_u32_exact_above_float24",
      "attrs": { "value": { "dtype": "uint32", "shape": [3], "values": [16777217, 16777217, 16777217] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "uint32", "shape": [3], "tolerance": 0 } }
    },
    {
      "name": "fill_rank0_scalar_output",
      "attrs": { "value_float": -3.75 },
      "inputs": {},
      "outputs": { "output": { "dtype": "float32", "shape": [], "tolerance": 0.000001 } }
    },
    {
      "name": "fill_zero_sized_output_noop",
      "attrs": { "value": { "dtype": "float32", "shape": [2, 0, 3], "values": [] } },
      "inputs": {},
      "outputs": { "output": { "dtype": "float32", "shape": [2, 0, 3], "tolerance": 0 } }
    },
    {
      "name": "onnx_backend_constant",
      "provenance": { "source": "cmake/external/onnx/onnx/backend/test/data/node/test_constant" },
      "attrs": {
        "value": {
          "dtype": "float32",
          "shape": [5, 5],
          "values": [1.764052391052246, 0.40015721321105957, 0.978738009929657, 2.2408931255340576, 1.8675580024719238, -0.9772778749465942, 0.9500884413719177, -0.15135720372200012, -0.10321885347366333, 0.4105985164642334, 0.14404356479644775, 1.4542734622955322, 0.7610377073287964, 0.12167501449584961, 0.44386324286460876, 0.3336743414402008, 1.4940791130065918, -0.2051582634449005, 0.3130677044391632, -0.8540957570075989, -2.5529897212982178, 0.653618574142456, 0.8644362092018127, -0.7421650290489197, 2.269754648208618]
        }
      },
      "outputs": { "output": { "dtype": "float32", "shape": [5, 5], "tolerance": 0 } }
    },
    {
      "name": "fill_float_positive_infinity",
      "attrs": { "value_float": "Infinity" },
      "inputs": {},
      "outputs": {
        "output": {
          "dtype": "float32",
          "shape": [],
          "tolerance": 0,
          "data": { "kind": "values", "values": ["Infinity"] }
        }
      }
    },
    {
      "name": "fill_float_negative_infinity",
      "attrs": { "value_float": "-Infinity" },
      "inputs": {},
      "outputs": {
        "output": {
          "dtype": "float32",
          "shape": [],
          "tolerance": 0,
          "data": { "kind": "values", "values": ["-Infinity"] }
        }
      }
    },
    {
      "name": "literal_values_dispatch_cliff_large_f32",
      "attrs": {
        "value_floats": [1, 2, 3, 4, 5, 6, 7, 8, 1, 2, 3, 4, 5, 6, 7, 8, 1, 2, 3, 4, 5, 6, 7, 8, 1, 2, 3, 4, 5, 6, 7, 8]
      },
      "inputs": {},
      "outputs": { "output": { "dtype": "float32", "shape": [32], "tolerance": 0 } }
    }
  ]
}