diff --git a/DESCRIPTION b/DESCRIPTION
index d4f286e..c1c6bb1 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -63,6 +63,7 @@ Roxygen: list(markdown = TRUE)
 RoxygenNote: 7.3.3
 Collate: 
     'bound_prediction.R'
+    'butcher.R'
     'cal-apply-binary.R'
     'cal-apply-impl.R'
     'cal-apply-multi.R'
diff --git a/NEWS.md b/NEWS.md
index 7be95a0..95e2855 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -2,6 +2,8 @@
 
 * Add `required_pkgs()` methods to `int_conformal_cv()`, `int_conformal_full()`, `int_conformal_quantile()`,  and `int_conformal_split()`. (#190)
 
+* Add butcher methods to `int_conformal_cv()`, `int_conformal_full()`, `int_conformal_quantile()`,  and `int_conformal_split()`. (#194)
+
 # probably 1.1.1
 
 * Updated unit tests for new ggplot2 release (#180).
diff --git a/R/butcher.R b/R/butcher.R
new file mode 100644
index 0000000..a0d1a96
--- /dev/null
+++ b/R/butcher.R
@@ -0,0 +1,173 @@
+#' Butcher methods for conformal inteference intervals
+#'
+#' These methods allow you to use the butcher package to reduce the size of a
+#' conformal inference interval object. After calling `butcher::butcher()` on a
+#' conformal inference interval object, the only guarantee is that you will
+#' still be able to `predict()` from that conformal inference interval object.
+#' Other functions may not work as expected.
+#'
+#' @param x A conformal inference interval object.
+#' @param verbose Should information be printed about how much memory is freed
+#'   from butchering?
+#' @param ... Extra arguments possibly used by underlying methods.
+#'
+#' @keywords internal
+#' @name inf_conformal-butcher
+
+# int_conformal_full
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_call.int_conformal_full <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_call(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_ctrl.int_conformal_full <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_ctrl(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_data.int_conformal_full <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_data(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_env.int_conformal_full <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_env(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_fitted.int_conformal_full <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_fitted(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# int_conformal_split
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_call.int_conformal_split <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_call(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_ctrl.int_conformal_split <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_ctrl(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_data.int_conformal_split <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_data(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_env.int_conformal_split <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_env(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_fitted.int_conformal_split <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_fitted(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# int_conformal_quantile
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_call.int_conformal_quantile <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_call(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_ctrl.int_conformal_quantile <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_ctrl(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_data.int_conformal_quantile <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_data(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_env.int_conformal_quantile <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_env(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_fitted.int_conformal_quantile <- function(x, verbose = FALSE, ...) {
+  x$wflow <- butcher::axe_fitted(x$wflow, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# int_conformal_cv
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_call.int_conformal_cv <- function(x, verbose = FALSE, ...) {
+  x$models <- purrr::map(x$models, butcher::axe_call, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_ctrl.int_conformal_cv <- function(x, verbose = FALSE, ...) {
+  x$models <- purrr::map(x$models, butcher::axe_ctrl, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_data.int_conformal_cv <- function(x, verbose = FALSE, ...) {
+  x$models <- purrr::map(x$models, butcher::axe_data, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_env.int_conformal_cv <- function(x, verbose = FALSE, ...) {
+  x$models <- purrr::map(x$models, butcher::axe_env, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# @export - onLoad
+#' @rdname inf_conformal-butcher
+axe_fitted.int_conformal_cv <- function(x, verbose = FALSE, ...) {
+  x$models <- purrr::map(x$models, butcher::axe_fitted, verbose = verbose, ...)
+  add_butcher_class(x)
+}
+
+# ------------------------------------------------------------------------------
+
+# butcher:::add_butcher_class
+add_butcher_class <- function(x) {
+  if (!any(grepl("butcher", class(x)))) {
+    class(x) <- append(paste0("butchered_", rev(class(x))[1]), class(x))
+  }
+  x
+}
diff --git a/R/zzz.R b/R/zzz.R
index b252929..7a8c19f 100644
--- a/R/zzz.R
+++ b/R/zzz.R
@@ -1,4 +1,28 @@
 .onLoad <- function(libname, pkgname) {
   vctrs::s3_register("tune::collect_metrics", "cal_rset")
   vctrs::s3_register("tune::collect_predictions", "cal_rset")
+
+  vctrs::s3_register("butcher::axe_call", "int_conformal_cv")
+  vctrs::s3_register("butcher::axe_ctrl", "int_conformal_cv")
+  vctrs::s3_register("butcher::axe_data", "int_conformal_cv")
+  vctrs::s3_register("butcher::axe_env", "int_conformal_cv")
+  vctrs::s3_register("butcher::axe_fitted", "int_conformal_cv")
+
+  vctrs::s3_register("butcher::axe_call", "int_conformal_full")
+  vctrs::s3_register("butcher::axe_ctrl", "int_conformal_full")
+  vctrs::s3_register("butcher::axe_data", "int_conformal_full")
+  vctrs::s3_register("butcher::axe_env", "int_conformal_full")
+  vctrs::s3_register("butcher::axe_fitted", "int_conformal_full")
+
+  vctrs::s3_register("butcher::axe_call", "int_conformal_split")
+  vctrs::s3_register("butcher::axe_ctrl", "int_conformal_split")
+  vctrs::s3_register("butcher::axe_data", "int_conformal_split")
+  vctrs::s3_register("butcher::axe_env", "int_conformal_split")
+  vctrs::s3_register("butcher::axe_fitted", "int_conformal_split")
+
+  vctrs::s3_register("butcher::axe_call", "int_conformal_quantile")
+  vctrs::s3_register("butcher::axe_ctrl", "int_conformal_quantile")
+  vctrs::s3_register("butcher::axe_data", "int_conformal_quantile")
+  vctrs::s3_register("butcher::axe_env", "int_conformal_quantile")
+  vctrs::s3_register("butcher::axe_fitted", "int_conformal_quantile")
 }
diff --git a/man/inf_conformal-butcher.Rd b/man/inf_conformal-butcher.Rd
new file mode 100644
index 0000000..e235a62
--- /dev/null
+++ b/man/inf_conformal-butcher.Rd
@@ -0,0 +1,82 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/butcher.R
+\name{inf_conformal-butcher}
+\alias{inf_conformal-butcher}
+\alias{axe_call.int_conformal_full}
+\alias{axe_ctrl.int_conformal_full}
+\alias{axe_data.int_conformal_full}
+\alias{axe_env.int_conformal_full}
+\alias{axe_fitted.int_conformal_full}
+\alias{axe_call.int_conformal_split}
+\alias{axe_ctrl.int_conformal_split}
+\alias{axe_data.int_conformal_split}
+\alias{axe_env.int_conformal_split}
+\alias{axe_fitted.int_conformal_split}
+\alias{axe_call.int_conformal_quantile}
+\alias{axe_ctrl.int_conformal_quantile}
+\alias{axe_data.int_conformal_quantile}
+\alias{axe_env.int_conformal_quantile}
+\alias{axe_fitted.int_conformal_quantile}
+\alias{axe_call.int_conformal_cv}
+\alias{axe_ctrl.int_conformal_cv}
+\alias{axe_data.int_conformal_cv}
+\alias{axe_env.int_conformal_cv}
+\alias{axe_fitted.int_conformal_cv}
+\title{Butcher methods for conformal inteference intervals}
+\usage{
+\method{axe_call}{int_conformal_full}(x, verbose = FALSE, ...)
+
+\method{axe_ctrl}{int_conformal_full}(x, verbose = FALSE, ...)
+
+\method{axe_data}{int_conformal_full}(x, verbose = FALSE, ...)
+
+\method{axe_env}{int_conformal_full}(x, verbose = FALSE, ...)
+
+\method{axe_fitted}{int_conformal_full}(x, verbose = FALSE, ...)
+
+\method{axe_call}{int_conformal_split}(x, verbose = FALSE, ...)
+
+\method{axe_ctrl}{int_conformal_split}(x, verbose = FALSE, ...)
+
+\method{axe_data}{int_conformal_split}(x, verbose = FALSE, ...)
+
+\method{axe_env}{int_conformal_split}(x, verbose = FALSE, ...)
+
+\method{axe_fitted}{int_conformal_split}(x, verbose = FALSE, ...)
+
+\method{axe_call}{int_conformal_quantile}(x, verbose = FALSE, ...)
+
+\method{axe_ctrl}{int_conformal_quantile}(x, verbose = FALSE, ...)
+
+\method{axe_data}{int_conformal_quantile}(x, verbose = FALSE, ...)
+
+\method{axe_env}{int_conformal_quantile}(x, verbose = FALSE, ...)
+
+\method{axe_fitted}{int_conformal_quantile}(x, verbose = FALSE, ...)
+
+\method{axe_call}{int_conformal_cv}(x, verbose = FALSE, ...)
+
+\method{axe_ctrl}{int_conformal_cv}(x, verbose = FALSE, ...)
+
+\method{axe_data}{int_conformal_cv}(x, verbose = FALSE, ...)
+
+\method{axe_env}{int_conformal_cv}(x, verbose = FALSE, ...)
+
+\method{axe_fitted}{int_conformal_cv}(x, verbose = FALSE, ...)
+}
+\arguments{
+\item{x}{A conformal inference interval object.}
+
+\item{verbose}{Should information be printed about how much memory is freed
+from butchering?}
+
+\item{...}{Extra arguments possibly used by underlying methods.}
+}
+\description{
+These methods allow you to use the butcher package to reduce the size of a
+conformal inference interval object. After calling \code{butcher::butcher()} on a
+conformal inference interval object, the only guarantee is that you will
+still be able to \code{predict()} from that conformal inference interval object.
+Other functions may not work as expected.
+}
+\keyword{internal}
diff --git a/tests/testthat/test-butcher.R b/tests/testthat/test-butcher.R
new file mode 100644
index 0000000..27f8d51
--- /dev/null
+++ b/tests/testthat/test-butcher.R
@@ -0,0 +1,111 @@
+test_that("attaches the butcher class", {
+  skip_if_not_installed("butcher")
+  skip_if_not_installed("modeldata")
+
+  suppressPackageStartupMessages(library(workflows))
+  suppressPackageStartupMessages(library(modeldata))
+  suppressPackageStartupMessages(library(purrr))
+  suppressPackageStartupMessages(library(rsample))
+  suppressPackageStartupMessages(library(tune))
+  suppressPackageStartupMessages(library(parsnip))
+  suppressPackageStartupMessages(library(butcher))
+
+  set.seed(2)
+  sim_train <- sim_regression(500)
+  sim_cal <- sim_regression(200)
+
+  # We'll use a neural network model
+  mlp_spec <-
+    mlp(hidden_units = 5, penalty = 0.01) |>
+    set_mode("regression")
+
+  mlp_wflow <-
+    workflow() |>
+    add_model(mlp_spec) |>
+    add_formula(outcome ~ .)
+
+  fit <- fit(mlp_wflow, data = sim_train)
+
+  c_int <- int_conformal_split(fit, sim_cal)
+  c_int <- butcher(c_int)
+  expect_s3_class(c_int, "butchered_int_conformal_split")
+
+  c_int <- int_conformal_full(fit, sim_cal)
+  c_int <- butcher(c_int)
+  expect_s3_class(c_int, "butchered_int_conformal_full")
+
+  c_int <- int_conformal_quantile(fit, sim_cal, sim_cal)
+  c_int <- butcher(c_int)
+  expect_s3_class(c_int, "butchered_int_conformal_quantile")
+
+  ctrl <- control_resamples(save_pred = TRUE, extract = I)
+
+  res <- mlp_wflow |>
+    fit_resamples(resamples = vfold_cv(sim_train, v = 2), control = ctrl)
+
+  fit <- int_conformal_cv(res)
+
+  c_int <- int_conformal_cv(res)
+  c_int <- butcher(c_int)
+  expect_s3_class(c_int, "butchered_int_conformal_cv")
+})
+
+test_that("butcher works", {
+  skip_if_not_installed("butcher")
+  skip_if_not_installed("modeldata")
+
+  suppressPackageStartupMessages(library(workflows))
+  suppressPackageStartupMessages(library(modeldata))
+  suppressPackageStartupMessages(library(purrr))
+  suppressPackageStartupMessages(library(rsample))
+  suppressPackageStartupMessages(library(tune))
+  suppressPackageStartupMessages(library(parsnip))
+  suppressPackageStartupMessages(library(butcher))
+
+  set.seed(2)
+  sim_train <- sim_regression(500)
+  sim_cal <- sim_regression(200)
+
+  # We'll use a neural network model
+  mlp_spec <-
+    mlp(hidden_units = 5, penalty = 0.01) |>
+    set_mode("regression")
+
+  mlp_wflow <-
+    workflow() |>
+    add_model(mlp_spec) |>
+    add_formula(outcome ~ .)
+
+  fit <- fit(mlp_wflow, data = sim_train)
+
+  c_int <- int_conformal_split(fit, sim_cal)
+  expect_identical(
+    butcher(c_int)$wflow,
+    butcher(c_int$wflow)
+  )
+
+  c_int <- int_conformal_full(fit, sim_cal)
+  expect_identical(
+    butcher(c_int)$wflow,
+    butcher(c_int$wflow)
+  )
+
+  c_int <- int_conformal_quantile(fit, sim_cal, sim_cal)
+  expect_identical(
+    butcher(c_int)$wflow,
+    butcher(c_int$wflow)
+  )
+
+  ctrl <- control_resamples(save_pred = TRUE, extract = I)
+
+  res <- mlp_wflow |>
+    fit_resamples(resamples = vfold_cv(sim_train, v = 2), control = ctrl)
+
+  fit <- int_conformal_cv(res)
+
+  c_int <- int_conformal_cv(res)
+  expect_identical(
+    butcher(c_int)$models,
+    lapply(c_int$models, butcher)
+  )
+})