diff --git a/src/nlp_expr.jl b/src/nlp_expr.jl
index b76cbcd706b..8ec71d21cf8 100644
--- a/src/nlp_expr.jl
+++ b/src/nlp_expr.jl
@@ -354,13 +354,13 @@ end
 
 # Multivariate operators
 
-# The multivariate operators in MOI are +, -, *, ^, /, ifelse, atan
+# The multivariate operators in MOI are +, -, *, ^, /, ifelse, atan, min, max
 #
 # However, ifelse is a builtin, so we can't add methods to it.
 
 # We need only very generic fallbacks for these, because all other cases are
 # caught with more specific methods.
-for f in (:+, :-, :*, :^, :/, :atan)
+for f in (:+, :-, :*, :^, :/, :atan, :min, :max)
     op = Meta.quot(f)
     @eval begin
         function Base.$(f)(x::AbstractJuMPScalar, y::_Constant)
diff --git a/test/test_nlp_expr.jl b/test/test_nlp_expr.jl
index bd309f4c1d8..386fca80097 100644
--- a/test/test_nlp_expr.jl
+++ b/test/test_nlp_expr.jl
@@ -926,4 +926,20 @@ function test_generic_nonlinear_expr_infer_variable_type()
     return
 end
 
+function test_operator_min()
+    model = Model()
+    @variable(model, x)
+    @test isequal_canonical(min(x, 1), NonlinearExpr(:min, Any[x, 1.0]))
+    @test isequal_canonical(min(1, x, x^2), min(min(1.0, x), x^2))
+    return
+end
+
+function test_operator_max()
+    model = Model()
+    @variable(model, x)
+    @test isequal_canonical(max(x, 1), NonlinearExpr(:max, Any[x, 1.0]))
+    @test isequal_canonical(max(1, x, x^2), max(max(1.0, x), x^2))
+    return
+end
+
 end  # module