Add basic benchmarks for model equality (#13061)

Author: Viicos

tests/benchmarks/test_model_eq.py

@@ -0,0 +1,38 @@
+from pydantic import BaseModel, create_model
+
+ModelNoExtra = create_model(
+    'ModelNoExtra',
+    __config__={'extra': 'forbid'},
+    **{f'f{i}': (int, i) for i in range(30)},  # pyright: ignore
+)
+
+ModelExtra = create_model(
+    'ModelNoExtra',
+    __config__={'extra': 'allow'},
+    **{f'f{i}': (int, i) for i in range(30)},  # pyright: ignore
+)
+
+
+def model_eq(m1: BaseModel, m2: BaseModel) -> bool:
+    return m1 == m2
+
+
+def test_model_eq_extra_forbid(benchmark) -> None:
+    m1 = ModelNoExtra()
+    m2 = ModelNoExtra()
+
+    benchmark(model_eq, m1, m2)
+
+
+def test_model_eq_extra_allow_no_extra(benchmark) -> None:
+    m1 = ModelExtra()
+    m2 = ModelExtra()
+
+    benchmark(model_eq, m1, m2)
+
+
+def test_model_eq_extra_allow_extra(benchmark) -> None:
+    m1 = ModelExtra(extra1='test', extra2='other', extra3=2)
+    m2 = ModelExtra(extra1='test', extra2='other', extra3=2)
+
+    benchmark(model_eq, m1, m2)