diff --git a/crates/typst-library/src/foundations/calc.rs b/crates/typst-library/src/foundations/calc.rs
index f12ca74ca..e7eb14055 100644
--- a/crates/typst-library/src/foundations/calc.rs
+++ b/crates/typst-library/src/foundations/calc.rs
@@ -50,6 +50,7 @@ pub fn module() -> Module {
     scope.define_func::<div_euclid>();
     scope.define_func::<rem_euclid>();
     scope.define_func::<quo>();
+    scope.define_func::<norm>();
     scope.define("inf", f64::INFINITY);
     scope.define("pi", std::f64::consts::PI);
     scope.define("tau", std::f64::consts::TAU);
@@ -1056,6 +1057,38 @@ pub fn quo(
     floor(divided).at(span)
 }
 
+/// Calculates the p-norm of a sequence of values.
+///
+/// ```example
+/// #calc.norm(1, 2, -3, 0.5) \
+/// #calc.norm(p: 3, 1, 2)
+/// ```
+#[func(title = "𝑝-Norm")]
+pub fn norm(
+    /// The p value to calculate the p-norm of.
+    #[named]
+    #[default(Spanned::new(2.0, Span::detached()))]
+    p: Spanned<f64>,
+    /// The sequence of values from which to calculate the p-norm.
+    /// Returns `0.0` if empty.
+    #[variadic]
+    values: Vec<f64>,
+) -> SourceResult<f64> {
+    if p.v <= 0.0 {
+        bail!(p.span, "p must be greater than zero");
+    }
+
+    // Create an iterator over the absolute values.
+    let abs = values.into_iter().map(f64::abs);
+
+    Ok(if p.v.is_infinite() {
+        // When p is infinity, the p-norm is the maximum of the absolute values.
+        abs.max_by(|a, b| a.total_cmp(b)).unwrap_or(0.0)
+    } else {
+        abs.map(|v| v.powf(p.v)).sum::<f64>().powf(1.0 / p.v)
+    })
+}
+
 /// A value which can be passed to functions that work with integers and floats.
 #[derive(Debug, Copy, Clone)]
 pub enum Num {
diff --git a/tests/suite/foundations/calc.typ b/tests/suite/foundations/calc.typ
index 5726dafa1..23fdea89c 100644
--- a/tests/suite/foundations/calc.typ
+++ b/tests/suite/foundations/calc.typ
@@ -368,3 +368,19 @@
 // Error: 2-37 cannot apply this operation to a decimal and a float
 // Hint: 2-37 if loss of precision is acceptable, explicitly cast the decimal to a float with `float(value)`
 #calc.clamp(decimal("10"), 5.5, 6.6)
+
+--- calc-norm ---
+#test(calc.norm(1, 2, -3, 0.5), calc.sqrt(14.25))
+#test(calc.norm(3, 4), 5.0)
+#test(calc.norm(3, 4), 5.0)
+#test(calc.norm(), 0.0)
+#test(calc.norm(p: 3, 1, -2), calc.pow(9, 1/3))
+#test(calc.norm(p: calc.inf, 1, -2), 2.0)
+
+--- calc-norm-negative-p ---
+// Error: 15-17 p must be greater than zero
+#calc.norm(p: -1, 1)
+
+--- calc-norm-expected-float ---
+// Error: 12-15 expected float, found ratio
+#calc.norm(10%)