diff --git a/src/uu/seq/src/seq.rs b/src/uu/seq/src/seq.rs
index 91dd091c4..777c77e1a 100644
--- a/src/uu/seq/src/seq.rs
+++ b/src/uu/seq/src/seq.rs
@@ -7,10 +7,11 @@ use std::ffi::OsString;
 use std::io::{stdout, BufWriter, ErrorKind, Write};
 
 use clap::{Arg, ArgAction, Command};
-use num_traits::{ToPrimitive, Zero};
+use num_traits::Zero;
 
 use uucore::error::{FromIo, UResult};
-use uucore::format::{num_format, sprintf, ExtendedBigDecimal, Format, FormatArgument};
+use uucore::format::num_format::FloatVariant;
+use uucore::format::{num_format, ExtendedBigDecimal, Format};
 use uucore::{format_usage, help_about, help_usage};
 
 mod error;
@@ -140,26 +141,52 @@ pub fn uumain(args: impl uucore::Args) -> UResult<()> {
         }
     };
 
-    let padding = first
-        .num_integral_digits
-        .max(increment.num_integral_digits)
-        .max(last.num_integral_digits);
-
     let precision = select_precision(first_precision, increment_precision, last_precision);
 
-    let format = options
-        .format
-        .map(Format::<num_format::Float, &ExtendedBigDecimal>::parse)
-        .transpose()?;
+    // If a format was passed on the command line, use that.
+    // If not, use some default format based on parameters precision.
+    let format = match options.format {
+        Some(str) => Format::<num_format::Float, &ExtendedBigDecimal>::parse(str)?,
+        None => {
+            let padding = if options.equal_width {
+                let precision_value = precision.unwrap_or(0);
+                first
+                    .num_integral_digits
+                    .max(increment.num_integral_digits)
+                    .max(last.num_integral_digits)
+                    + if precision_value > 0 {
+                        precision_value + 1
+                    } else {
+                        0
+                    }
+            } else {
+                0
+            };
+
+            let formatter = match precision {
+                // format with precision: decimal floats and integers
+                Some(precision) => num_format::Float {
+                    variant: FloatVariant::Decimal,
+                    width: padding,
+                    alignment: num_format::NumberAlignment::RightZero,
+                    precision,
+                    ..Default::default()
+                },
+                // format without precision: hexadecimal floats
+                None => num_format::Float {
+                    variant: FloatVariant::Shortest,
+                    ..Default::default()
+                },
+            };
+            Format::from_formatter(formatter)
+        }
+    };
 
     let result = print_seq(
         (first.number, increment.number, last.number),
-        precision,
         &options.separator,
         &options.terminator,
-        options.equal_width,
-        padding,
-        format.as_ref(),
+        &format,
     );
     match result {
         Ok(()) => Ok(()),
@@ -218,84 +245,24 @@ fn done_printing<T: Zero + PartialOrd>(next: &T, increment: &T, last: &T) -> boo
     }
 }
 
-fn format_bigdecimal(value: &bigdecimal::BigDecimal) -> Option<String> {
-    let format_arguments = &[FormatArgument::Float(value.to_f64()?)];
-    let value_as_bytes = sprintf("%g", format_arguments).ok()?;
-    String::from_utf8(value_as_bytes).ok()
-}
-
-/// Write a big decimal formatted according to the given parameters.
-fn write_value_float(
-    writer: &mut impl Write,
-    value: &ExtendedBigDecimal,
-    width: usize,
-    precision: Option<usize>,
-) -> std::io::Result<()> {
-    let value_as_str = match precision {
-        // format with precision: decimal floats and integers
-        Some(precision) => match value {
-            ExtendedBigDecimal::Infinity | ExtendedBigDecimal::MinusInfinity => {
-                format!("{value:>width$.precision$}")
-            }
-            _ => format!("{value:>0width$.precision$}"),
-        },
-        // format without precision: hexadecimal floats
-        None => match value {
-            ExtendedBigDecimal::BigDecimal(bd) => {
-                format_bigdecimal(bd).unwrap_or_else(|| "{value}".to_owned())
-            }
-            _ => format!("{value:>0width$}"),
-        },
-    };
-    write!(writer, "{value_as_str}")
-}
-
 /// Floating point based code path
 fn print_seq(
     range: RangeFloat,
-    precision: Option<usize>,
     separator: &str,
     terminator: &str,
-    pad: bool,
-    padding: usize,
-    format: Option<&Format<num_format::Float, &ExtendedBigDecimal>>,
+    format: &Format<num_format::Float, &ExtendedBigDecimal>,
 ) -> std::io::Result<()> {
     let stdout = stdout().lock();
     let mut stdout = BufWriter::new(stdout);
     let (first, increment, last) = range;
     let mut value = first;
-    let padding = if pad {
-        let precision_value = precision.unwrap_or(0);
-        padding
-            + if precision_value > 0 {
-                precision_value + 1
-            } else {
-                0
-            }
-    } else {
-        0
-    };
+
     let mut is_first_iteration = true;
     while !done_printing(&value, &increment, &last) {
         if !is_first_iteration {
             write!(stdout, "{separator}")?;
         }
-        // If there was an argument `-f FORMAT`, then use that format
-        // template instead of the default formatting strategy.
-        //
-        // TODO The `printf()` method takes a string as its second
-        // parameter but we have an `ExtendedBigDecimal`. In order to
-        // satisfy the signature of the function, we convert the
-        // `ExtendedBigDecimal` into a string. The `printf()`
-        // logic will subsequently parse that string into something
-        // similar to an `ExtendedBigDecimal` again before rendering
-        // it as a string and ultimately writing to `stdout`. We
-        // shouldn't have to do so much converting back and forth via
-        // strings.
-        match &format {
-            Some(f) => f.fmt(&mut stdout, &value)?,
-            None => write_value_float(&mut stdout, &value, padding, precision)?,
-        }
+        format.fmt(&mut stdout, &value)?;
         // TODO Implement augmenting addition.
         value = value + increment.clone();
         is_first_iteration = false;
diff --git a/src/uucore/src/lib/features/format/mod.rs b/src/uucore/src/lib/features/format/mod.rs
index e44ef4bc0..25a4449dc 100644
--- a/src/uucore/src/lib/features/format/mod.rs
+++ b/src/uucore/src/lib/features/format/mod.rs
@@ -310,12 +310,12 @@ pub fn sprintf<'a>(
     Ok(writer)
 }
 
-/// A parsed format for a single numerical value of type T
+/// A format for a single numerical value of type T
 ///
-/// This is used by `seq` and `csplit`. It can be constructed with [`Format::parse`]
-/// and can write a value with [`Format::fmt`].
+/// This is used by `seq` and `csplit`. It can be constructed with [`Format::from_formatter`]
+/// or [`Format::parse`] and can write a value with [`Format::fmt`].
 ///
-/// It can only accept a single specification without any asterisk parameters.
+/// [`Format::parse`] can only accept a single specification without any asterisk parameters.
 /// If it does get more specifications, it will return an error.
 pub struct Format<F: Formatter<T>, T> {
     prefix: Vec<u8>,
@@ -325,6 +325,15 @@ pub struct Format<F: Formatter<T>, T> {
 }
 
 impl<F: Formatter<T>, T> Format<F, T> {
+    pub fn from_formatter(formatter: F) -> Self {
+        Self {
+            prefix: Vec::<u8>::new(),
+            suffix: Vec::<u8>::new(),
+            formatter,
+            _marker: PhantomData,
+        }
+    }
+
     pub fn parse(format_string: impl AsRef<[u8]>) -> Result<Self, FormatError> {
         let mut iter = parse_spec_only(format_string.as_ref());