diff --git a/.github/workflows/CI.yml b/.github/workflows/CI.yml
index f796a66d7..353819673 100644
--- a/.github/workflows/CI.yml
+++ b/.github/workflows/CI.yml
@@ -9,6 +9,12 @@ on:
       - master
     tags: '*'
 
+concurrency:
+  # Skip intermediate builds: always.
+  # Cancel intermediate builds: only if it is a pull request build.
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: ${{ startsWith(github.ref, 'refs/pull/') }}
+
 jobs:
   test:
     name: ${{ matrix.suite }} - Julia ${{ matrix.version }} - ${{ matrix.os }} - ${{ matrix.arch }}
@@ -93,3 +99,29 @@ jobs:
       - uses: codecov/codecov-action@v2
         with:
           file: lcov.info
+
+  docs:
+    name: Documentation
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - uses: julia-actions/setup-julia@v1
+        with:
+          version: '1.6'
+      - run: |
+          julia --project=docs -e '
+            using Pkg
+            Pkg.develop(PackageSpec(path=pwd()))
+            Pkg.instantiate()'
+      - run: |
+          julia --color=yes --project=docs/ -e '
+            using Metalhead
+            # using Pkg; Pkg.activate("docs")
+            using Documenter
+            using Documenter: doctest
+            DocMeta.setdocmeta!(Metalhead, :DocTestSetup, :(using Metalhead); recursive=true)
+            doctest(Metalhead)'
+      - run: julia --project=docs docs/make.jl
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          DOCUMENTER_KEY: ${{ secrets.DOCUMENTER_KEY }}
diff --git a/.github/workflows/Publish.yml b/.github/workflows/Publish.yml
deleted file mode 100644
index 5d59ed275..000000000
--- a/.github/workflows/Publish.yml
+++ /dev/null
@@ -1,33 +0,0 @@
-# Adapted from https://github.com/MichaelHatherly/Publish.jl
-name: Publish
-
-on:
-  push:
-    branches: 'master'
-
-jobs:
-  publish:
-    name: "Publish Documentation (dev)"
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v2
-      - uses: actions/checkout@v2
-        with:
-          ref: gh-pages
-          path: gh-pages
-      - uses: julia-actions/setup-julia@latest
-        with:
-          version: '1'
-      - name: Install dependencies
-        run: |
-          julia --color=yes --project=. -e 'using Pkg; Pkg.instantiate()'
-      - name: Build and deploy
-        run: |
-          cd gh-pages
-          julia --color=yes --project=../docs -e "using Pkg; Pkg.instantiate()"
-          julia --color=yes --project=../docs -e 'include("../docs/make.jl"); build_and_deploy("dev")'
-          git config user.name github-actions
-          git config user.email github-actions@github.com
-          git add .
-          git commit -m "(dev) Built by Publish.jl."
-          git push
diff --git a/.github/workflows/PublishStable.yml b/.github/workflows/PublishStable.yml
deleted file mode 100644
index d18b27482..000000000
--- a/.github/workflows/PublishStable.yml
+++ /dev/null
@@ -1,32 +0,0 @@
-name: PublishStable
-
-on:
-  push:
-    tags:
-      - '*'
-
-jobs:
-  publish:
-    name: "Publish Documentation (stable)"
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v2
-      - uses: actions/checkout@v2
-        with:
-          ref: gh-pages
-          path: gh-pages
-      - uses: julia-actions/setup-julia@latest
-        with:
-          version: '1.6'
-      - name: Install dependencies
-        run: julia --color=yes --project=. -e 'using Pkg; Pkg.instantiate()'
-      - name: Build and deploy
-        run: |
-          cd gh-pages
-          julia --color=yes --project=../docs -e "using Pkg; Pkg.instantiate()"
-          julia --color=yes --project=../docs -e 'include("../docs/make.jl"); build_and_deploy("stable")'
-          git config user.name github-actions
-          git config user.email github-actions@github.com
-          git add .
-          git commit -m "(stable) Built by Publish.jl."
-          git push
diff --git a/.github/workflows/clean_preview.yml b/.github/workflows/clean_preview.yml
new file mode 100644
index 000000000..25946efc3
--- /dev/null
+++ b/.github/workflows/clean_preview.yml
@@ -0,0 +1,27 @@
+# from https://github.com/CliMA/ClimaTimeSteppers.jl
+name: Doc Preview Cleanup
+
+on:
+  pull_request:
+    types: [closed]
+
+jobs:
+  doc-preview-cleanup:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout gh-pages branch
+        uses: actions/checkout@v2
+        with:
+          ref: gh-pages
+      - name: Delete preview and history + push changes
+        run: |
+            if [ -d "previews/PR$PRNUM" ]; then
+              git config user.name "Documenter.jl"
+              git config user.email "documenter@juliadocs.github.io"
+              git rm -rf "previews/PR$PRNUM"
+              git commit -m "delete preview"
+              git branch gh-pages-new $(echo "delete history" | git commit-tree HEAD^{tree})
+              git push --force origin gh-pages-new:gh-pages
+            fi
+        env:
+            PRNUM: ${{ github.event.number }}
diff --git a/.github/workflows/pr_comment.yml b/.github/workflows/pr_comment.yml
new file mode 100644
index 000000000..0106e9029
--- /dev/null
+++ b/.github/workflows/pr_comment.yml
@@ -0,0 +1,14 @@
+name: pr_comment
+on:
+  pull_request:
+    types: [labeled]
+jobs:
+  pr_comment:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Create PR comment
+        if: github.event_name == 'pull_request' && github.repository == github.event.pull_request.head.repo.full_name && github.event.label.name == 'documentation' # if this is a pull request build AND the pull request is NOT made from a fork
+        uses: thollander/actions-comment-pull-request@71efef56b184328c7ef1f213577c3a90edaa4aff
+        with:
+          message: 'Once the documentation build has completed, you can preview any updated documentation at this URL: https://fluxml.ai/Metalhead.jl/previews/PR${{ github.event.number }}/'
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
diff --git a/README.md b/README.md
index 487579e25..da77712d1 100644
--- a/README.md
+++ b/README.md
@@ -9,36 +9,36 @@
 ## Installation
 
 ```julia
-]add Metalhead
+julia> ]add Metalhead
 ```
 
 ## Available models
 
-| Model Name                                       | Function                                                                                  | Pre-trained? |
-|:-------------------------------------------------|:------------------------------------------------------------------------------------------|:------------:|
-| [VGG](https://arxiv.org/abs/1409.1556)           | [`VGG`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.VGG.html)                 | Y (w/o BN)   |
-| [ResNet](https://arxiv.org/abs/1512.03385)       | [`ResNet`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.ResNet.html)           | Y            |
-| [WideResNet](https://arxiv.org/abs/1605.07146)   | [`WideResNet`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.WideResNet.html) | Y            |
-| [GoogLeNet](https://arxiv.org/abs/1409.4842)     | [`GoogLeNet`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.GoogLeNet.html)     | N            |
-| [Inception-v3](https://arxiv.org/abs/1512.00567) | [`Inceptionv3`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.Inceptionv3.html)   | N            |
-| [Inception-v4](https://arxiv.org/abs/1602.07261) | [`Inceptionv4`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.Inceptionv4.html)   | N            |
-| [InceptionResNet-v2](https://arxiv.org/abs/1602.07261) | [`Inceptionv3`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.InceptionResNetv2.html)   | N            |
-| [SqueezeNet](https://arxiv.org/abs/1602.07360)   | [`SqueezeNet`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.SqueezeNet.html)   | Y            |
-| [DenseNet](https://arxiv.org/abs/1608.06993)     | [`DenseNet`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.DenseNet.html)       | N            |
-| [ResNeXt](https://arxiv.org/abs/1611.05431)      | [`ResNeXt`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.ResNeXt.html)         | Y            |
-| [MobileNetv1](https://arxiv.org/abs/1704.04861)  | [`MobileNetv1`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.MobileNetv1.html) | N            |
-| [MobileNetv2](https://arxiv.org/abs/1801.04381)  | [`MobileNetv2`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.MobileNetv2.html) | N            |
-| [MobileNetv3](https://arxiv.org/abs/1905.02244)  | [`MobileNetv3`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.MobileNetv3.html) | N            |
-| [EfficientNet](https://arxiv.org/abs/1905.11946) | [`EfficientNet`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.EfficientNet.html) | N            |
-| [MLPMixer](https://arxiv.org/pdf/2105.01601)     | [`MLPMixer`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.MLPMixer.html)       | N            |
-| [ResMLP](https://arxiv.org/abs/2105.03404)       | [`ResMLP`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.ResMLP.html)           | N            |
-| [gMLP](https://arxiv.org/abs/2105.08050)         | [`gMLP`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.gMLP.html)               | N            |
-| [ViT](https://arxiv.org/abs/2010.11929)          | [`ViT`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.ViT.html)                 | N            |
-| [ConvNeXt](https://arxiv.org/abs/2201.03545)     | [`ConvNeXt`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.ConvNeXt.html)       | N            |
-| [ConvMixer](https://arxiv.org/abs/2201.09792)    | [`ConvMixer`](https://fluxml.ai/Metalhead.jl/dev/docstrings/Metalhead.ConvMixer.html)     | N            |
-
-To contribute new models, see our [contributing docs](https://fluxml.ai/Metalhead.jl/dev/docs/dev-guide/contributing.html).
+| Model Name                                       | Function                                                                                       | Pre-trained? |
+|:-------------------------------------------------|:-----------------------------------------------------------------------------------------------|:------------:|
+| [VGG](https://arxiv.org/abs/1409.1556)           | [`VGG`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.VGG)                   | Y (w/o BN)   |
+| [ResNet](https://arxiv.org/abs/1512.03385)       | [`ResNet`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.ResNet)             | Y            |
+| [WideResNet](https://arxiv.org/abs/1605.07146)   | [`WideResNet`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.WideResNet)     | Y            |
+| [GoogLeNet](https://arxiv.org/abs/1409.4842)     | [`GoogLeNet`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.GoogLeNet)       | N            |
+| [Inception-v3](https://arxiv.org/abs/1512.00567) | [`Inceptionv3`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.Inceptionv3)   | N            |
+| [Inception-v4](https://arxiv.org/abs/1602.07261) | [`Inceptionv4`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.Inceptionv4)   | N            |
+| [InceptionResNet-v2](https://arxiv.org/abs/1602.07261) | [`Inceptionv3`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.InceptionResNetv2) | N            |
+| [SqueezeNet](https://arxiv.org/abs/1602.07360)   | [`SqueezeNet`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.SqueezeNet)     | Y            |
+| [DenseNet](https://arxiv.org/abs/1608.06993)     | [`DenseNet`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.DenseNet)         | N            |
+| [ResNeXt](https://arxiv.org/abs/1611.05431)      | [`ResNeXt`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.ResNeXt)           | Y            |
+| [MobileNetv1](https://arxiv.org/abs/1704.04861)  | [`MobileNetv1`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.MobileNetv1)   | N            |
+| [MobileNetv2](https://arxiv.org/abs/1801.04381)  | [`MobileNetv2`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.MobileNetv2)   | N            |
+| [MobileNetv3](https://arxiv.org/abs/1905.02244)  | [`MobileNetv3`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.MobileNetv3)   | N            |
+| [EfficientNet](https://arxiv.org/abs/1905.11946) | [`EfficientNet`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.EfficientNet) | N            |
+| [MLPMixer](https://arxiv.org/pdf/2105.01601)     | [`MLPMixer`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.MLPMixer)         | N            |
+| [ResMLP](https://arxiv.org/abs/2105.03404)       | [`ResMLP`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.ResMLP)             | N            |
+| [gMLP](https://arxiv.org/abs/2105.08050)         | [`gMLP`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.gMLP)                 | N            |
+| [ViT](https://arxiv.org/abs/2010.11929)          | [`ViT`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.ViT)                   | N            |
+| [ConvNeXt](https://arxiv.org/abs/2201.03545)     | [`ConvNeXt`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.ConvNeXt)         | N            |
+| [ConvMixer](https://arxiv.org/abs/2201.09792)    | [`ConvMixer`](https://fluxml.ai/Metalhead.jl/stable/api/reference.html#Metalhead.ConvMixer)       | N            |
+
+To contribute new models, see our [contributing docs](https://fluxml.ai/Metalhead.jl/latest/contributing/).
 
 ## Getting Started
 
-You can find the Metalhead.jl getting started guide [here](https://fluxml.ai/Metalhead.jl/dev/docs/tutorials/quickstart.html).
+You can find the Metalhead.jl getting started guide [here](https://fluxml.ai/Metalhead.jl/latest/tutorials/quickstart/).
diff --git a/docs/.gitignore b/docs/.gitignore
new file mode 100644
index 000000000..a303fff20
--- /dev/null
+++ b/docs/.gitignore
@@ -0,0 +1,2 @@
+build/
+site/
diff --git a/docs/Project.toml b/docs/Project.toml
index e5e6979e1..6335fa0e9 100644
--- a/docs/Project.toml
+++ b/docs/Project.toml
@@ -1,6 +1,8 @@
 [deps]
+Artifacts = "56f22d72-fd6d-98f1-02f0-08ddc0907c33"
 DataAugmentation = "88a5189c-e7ff-4f85-ac6b-e6158070f02e"
+Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
 Flux = "587475ba-b771-5e3f-ad9e-33799f191a9c"
 Images = "916415d5-f1e6-5110-898d-aaa5f9f070e0"
-Metalhead = "dbeba491-748d-5e0e-a39e-b530a07fa0cc"
-Publish = "f065f642-d108-4f50-8aa5-6749150a895a"
+LazyArtifacts = "4af54fe1-eca0-43a8-85a7-787d91b784e3"
+OneHotArrays = "0b1bfda6-eb8a-41d2-88d8-f5af5cad476f"
diff --git a/docs/make.jl b/docs/make.jl
index f5d29f7e9..72fb486c9 100644
--- a/docs/make.jl
+++ b/docs/make.jl
@@ -1,17 +1,26 @@
-using Pkg
+using Documenter, Metalhead, Artifacts, LazyArtifacts, Images, OneHotArrays, DataAugmentation, Flux
 
-Pkg.develop(; path = "..")
+DocMeta.setdocmeta!(Metalhead, :DocTestSetup, :(using Metalhead); recursive = true)
 
-using Publish
-using Artifacts, LazyArtifacts
-using Metalhead
+makedocs(modules = [Metalhead, Artifacts, LazyArtifacts, Images, OneHotArrays, DataAugmentation, Flux],
+         sitename = "Metalhead.jl",
+         doctest = false,
+         pages = ["Home" => "index.md",
+                  "Tutorials" => [
+                      "tutorials/quickstart.md",
+                   ],
+                  "Developer guide" => "contributing.md",
+                  "API reference" => [
+                      "api/reference.md",
+                   ],
+                 ],
+         format = Documenter.HTML(
+              canonical = "https://fluxml.ai/Metalhead.jl/stable/",
+            #   analytics = "UA-36890222-9",
+              assets = ["assets/flux.css"],
+              prettyurls = get(ENV, "CI", nothing) == "true"),
+        )
 
-# override default theme
-cp(artifact"flux-theme", "../_flux-theme"; force = true)
-
-p = Publish.Project(Metalhead)
-
-function build_and_deploy(label)
-    rm(label; recursive = true, force = true)
-    return deploy(Metalhead; root = "/Metalhead.jl", label = label)
-end
+deploydocs(repo = "github.com/FluxML/Metalhead.jl.git",
+           target = "build",
+           push_preview = true)
diff --git a/docs/serve.jl b/docs/serve.jl
deleted file mode 100644
index bf4a51179..000000000
--- a/docs/serve.jl
+++ /dev/null
@@ -1,17 +0,0 @@
-using Pkg
-
-Pkg.develop(; path = "..")
-
-using Revise
-using Publish
-using Artifacts, LazyArtifacts
-
-using Metalhead
-
-# override default theme
-cp(artifact"flux-theme", "../_flux-theme"; force = true)
-
-p = Publish.Project(Metalhead)
-
-# serve documentation
-serve(Metalhead)
diff --git a/docs/src/api/reference.md b/docs/src/api/reference.md
new file mode 100644
index 000000000..1699e81fe
--- /dev/null
+++ b/docs/src/api/reference.md
@@ -0,0 +1,17 @@
+# API Reference
+
+The API reference of `Metalhead.jl`.
+
+**Note**: This page is still in progress.
+
+```@autodocs
+Modules = [Metalhead]
+```
+
+```@docs
+Metalhead.create_classifier
+Metalhead.squeeze_excite
+Metalhead.LayerScale
+Metalhead.DropBlock
+Metalhead.StochasticDepth
+```
diff --git a/docs/src/assets/flux.css b/docs/src/assets/flux.css
new file mode 100644
index 000000000..541ead5fe
--- /dev/null
+++ b/docs/src/assets/flux.css
@@ -0,0 +1,113 @@
+@import url('https://fonts.googleapis.com/css?family=Lato:400,400i');
+
+body {
+  font-family: Lato, "Segoe UI",Roboto,"Helvetica Neue",Arial,sans-serif;
+}
+
+nav.toc {
+  padding-top: 0;
+  background: rgb(240, 240, 240);
+  line-height: 2em;
+  cursor: default;
+  user-select: none;
+}
+
+h1+h2 {
+  margin-top: 0;
+}
+
+/* Green banner in ToC */
+nav.toc > h1 {
+  margin-top: 0;
+  padding-top: 0.4em;
+  padding-bottom: 0.5em;
+  border-bottom: 5px solid white;
+  box-shadow: 0px -2px 5px rgb(60,60,60);
+  margin-bottom: 0.5em;
+  background: rgb(60, 150, 60);
+
+  font-style: italic;
+  font-weight: normal;
+  font-size: 50pt;
+  text-transform: lowercase;
+  text-shadow: 2px 2px 5px rgba(0,0,0,0.2);
+  color: white;
+}
+
+/* Reduce ToC font size */
+.toctext {
+  font-size: 10pt;
+}
+
+/* Fade out non-clickable ToC headers */
+nav.toc ul span.toctext {
+  color: rgb(180, 180, 180);
+}
+
+nav.toc ul .toctext {
+  color: rgb(100, 100, 100);
+}
+
+nav.toc ul a.toctext:hover {
+  color: inherit;
+  background: rgb(220, 220, 220);
+  cursor: default;
+}
+
+nav.toc li.current > .toctext {
+  background: linear-gradient(90deg, rgb(245,245,245) 0%, white 90%);
+  font-weight: normal;
+}
+
+nav.toc ul.internal li.toplevel {
+  font-weight: normal;
+}
+
+/* Content */
+
+article { max-width: none; }
+
+article > p, article > ul {
+  max-width: 45em;
+}
+
+/* Links */
+a, a:visited { color: rgb(0, 120, 0); }
+article p a { border-bottom: 1px solid rgb(200, 230, 200); }
+a:hover, a:visited:hover { color: rgb(0, 80, 0); }
+
+/* Article Links */
+article p a { border-bottom: 1px solid rgb(200, 230, 200); }
+article p a:hover, article a:visited:hover { color: rgb(0, 120, 0); }
+article p a:hover { border-bottom: 1px solid rgb(150, 200, 150); }
+
+/* Doctstrings */
+article section.docstring {
+  padding: 0.5em 0;
+  border-left: none;
+  border-right: none;
+  border-bottom: none;
+}
+
+/* Code */
+
+article pre, article p > code {
+  background: rgb(245, 250, 245);
+}
+
+article pre {
+  border: none;
+  max-width: none;
+  padding: 1em;
+  border-radius: 10px 0px 0px 10px;
+  margin-left: -1em;
+  margin-right: -2em;
+}
+
+.hljs-comment {
+  font-style: italic;
+}
+
+.hljs-number {
+  color: rgb(0, 150, 150);
+}
diff --git a/docs/dev-guide/contributing.md b/docs/src/contributing.md
similarity index 100%
rename from docs/dev-guide/contributing.md
rename to docs/src/contributing.md
diff --git a/docs/src/index.md b/docs/src/index.md
new file mode 100644
index 000000000..920cb8ba0
--- /dev/null
+++ b/docs/src/index.md
@@ -0,0 +1,44 @@
+```@meta
+CurrentModule = Metalhead
+```
+
+# Metalhead
+
+[Metalhead.jl](https://github.com/FluxML/Metalhead.jl) provides standard machine learning vision models for use with [Flux.jl](https://fluxml.ai). The architectures in this package make use of pure Flux layers, and they represent the best-practices for creating modules like residual blocks, inception blocks, etc. in Flux. Metalhead also provides some building blocks for more complex models in the Layers module.
+
+## Installation
+
+```julia
+julia> ]add Metalhead
+```
+
+## Available models
+
+| Model Name                                             | Function                    | Pre-trained? |
+|:-------------------------------------------------------|:----------------------------|:------------:|
+| [VGG](https://arxiv.org/abs/1409.1556)                 | [`VGG`](@ref)               | Y (w/o BN)   |
+| [ResNet](https://arxiv.org/abs/1512.03385)             | [`ResNet`](@ref)            | Y            |
+| [WideResNet](https://arxiv.org/abs/1605.07146)         | [`WideResNet`](@ref)        | Y            |
+| [GoogLeNet](https://arxiv.org/abs/1409.4842)           | [`GoogLeNet`](@ref)         | N            |
+| [Inception-v3](https://arxiv.org/abs/1512.00567)       | [`Inceptionv3`](@ref)       | N            |
+| [Inception-v4](https://arxiv.org/abs/1602.07261)       | [`Inceptionv4`](@ref)       | N            |
+| [InceptionResNet-v2](https://arxiv.org/abs/1602.07261) | [`Inceptionv3`](@ref)       | N            |
+| [SqueezeNet](https://arxiv.org/abs/1602.07360)         | [`SqueezeNet`](@ref)        | Y            |
+| [DenseNet](https://arxiv.org/abs/1608.06993)           | [`DenseNet`](@ref)          | N            |
+| [ResNeXt](https://arxiv.org/abs/1611.05431)            | [`ResNeXt`](@ref)           | Y            |
+| [MobileNetv1](https://arxiv.org/abs/1704.04861)        | [`MobileNetv1`](@ref)       | N            |
+| [MobileNetv2](https://arxiv.org/abs/1801.04381)        | [`MobileNetv2`](@ref)       | N            |
+| [MobileNetv3](https://arxiv.org/abs/1905.02244)        | [`MobileNetv3`](@ref)       | N            |
+| [EfficientNet](https://arxiv.org/abs/1905.11946)       | [`EfficientNet`](@ref)      | N            |
+| [MLPMixer](https://arxiv.org/pdf/2105.01601)           | [`MLPMixer`](@ref)          | N            |
+| [ResMLP](https://arxiv.org/abs/2105.03404)             | [`ResMLP`](@ref)            | N            |
+| [gMLP](https://arxiv.org/abs/2105.08050)               | [`gMLP`](@ref)              | N            |
+| [ViT](https://arxiv.org/abs/2010.11929)                | [`ViT`](@ref)               | N            |
+| [ConvNeXt](https://arxiv.org/abs/2201.03545)           | [`ConvNeXt`](@ref)          | N            |
+| [ConvMixer](https://arxiv.org/abs/2201.09792)          | [`ConvMixer`](@ref)         | N            |
+
+To contribute new models, see our [contributing docs](@ref Contributing-to-Metalhead.jl).
+
+## Getting Started
+
+You can find the Metalhead.jl getting started guide [here](@ref Quickstart).
diff --git a/docs/tutorials/quickstart.md b/docs/src/tutorials/quickstart.md
similarity index 82%
rename from docs/tutorials/quickstart.md
rename to docs/src/tutorials/quickstart.md
index cfb99eb72..a00854627 100644
--- a/docs/tutorials/quickstart.md
+++ b/docs/src/tutorials/quickstart.md
@@ -1,12 +1,10 @@
 # Quickstart
 
-{cell=quickstart, display=false, output=false, results=false}
 ```julia
 using Flux, Metalhead
 ```
 
-Using a model from Metalhead is as simple as selecting a model from the table of [available models](@ref). For example, below we use the pre-trained ResNet-18 model.
-{cell=quickstart}
+Using a model from Metalhead is as simple as selecting a model from the table of [available models](@ref API-Reference). For example, below we use the pre-trained ResNet-18 model.
 ```julia
 using Flux, Metalhead
 
@@ -16,7 +14,6 @@ model = ResNet(18; pretrain = true)
 Now, we can use this model with Flux like any other model.
 
 First, let's check the accuracy on a test image from ImageNet.
-{cell=quickstart}
 ```julia
 using Images
 
@@ -24,9 +21,8 @@ using Images
 img = Images.load(download("https://cdn.pixabay.com/photo/2015/05/07/11/02/guitar-756326_960_720.jpg"));
 ```
 We'll use the popular [DataAugmentation.jl](https://github.com/lorenzoh/DataAugmentation.jl) library to crop our input image, convert it to a plain array, and normalize the pixels.
-{cell=quickstart}
 ```julia
-using DataAugmentation
+using DataAugmentation, OneHotArrays
 
 DATA_MEAN = (0.485, 0.456, 0.406)
 DATA_STD = (0.229, 0.224, 0.225)
@@ -39,13 +35,13 @@ data = apply(augmentations, Image(img)) |> itemdata
 # image net labels
 labels = readlines(download("https://raw.githubusercontent.com/pytorch/hub/master/imagenet_classes.txt"))
 
-Flux.onecold(model(Flux.unsqueeze(data, 4)), labels)
+onecold(model(Flux.unsqueeze(data, 4)), labels)
 ```
 
 Below, we train it on some randomly generated data.
 
 ```julia
-using Flux: onehotbatch
+using OneHotArrays: onehotbatch
 
 batchsize = 1
 data = [(rand(Float32, 224, 224, 3, batchsize), onehotbatch(rand(1:1000, batchsize), 1:1000))
diff --git a/src/convnets/alexnet.jl b/src/convnets/alexnet.jl
index 51cfbf029..fba6749e4 100644
--- a/src/convnets/alexnet.jl
+++ b/src/convnets/alexnet.jl
@@ -45,7 +45,7 @@ Create a `AlexNet`.
     
     `AlexNet` does not currently support pretrained weights.
 
-See also [`Metalhead.alexnet`](@ref).
+See also [`alexnet`](@ref).
 """
 struct AlexNet
     layers::Any
diff --git a/src/convnets/builders/mbconv.jl b/src/convnets/builders/mbconv.jl
index 7b68be300..2f2dadfdb 100644
--- a/src/convnets/builders/mbconv.jl
+++ b/src/convnets/builders/mbconv.jl
@@ -6,7 +6,7 @@
 
 Creates a block builder for `irblockfn` within a given stage.
 Note that this function is not intended to be called directly, but instead passed to
-[`mbconv_stage_builder`](@ref) which will return a builder over all stages.
+[`Metalhead.mbconv_stage_builder`](@ref) which will return a builder over all stages.
 Users wanting to provide a custom inverted residual block type can extend this
 function by defining `invresbuilder(::typeof(my_block), ...)`.
 """
diff --git a/src/convnets/convnext.jl b/src/convnets/convnext.jl
index 30b314556..a38840d0b 100644
--- a/src/convnets/convnext.jl
+++ b/src/convnets/convnext.jl
@@ -8,7 +8,7 @@ Creates a single block of ConvNeXt.
 
   - `planes`: number of input channels.
   - `stochastic_depth_prob`: Stochastic depth probability.
-  - `layerscale_init`: Initial value for [`LayerScale`](@ref)
+  - `layerscale_init`: Initial value for [`Metalhead.LayerScale`](@ref)
 """
 function convnextblock(planes::Integer, stochastic_depth_prob = 0.0,
                        layerscale_init = 1.0f-6)
diff --git a/src/convnets/squeezenet.jl b/src/convnets/squeezenet.jl
index 5d63a565d..1e90dd55c 100644
--- a/src/convnets/squeezenet.jl
+++ b/src/convnets/squeezenet.jl
@@ -63,7 +63,7 @@ Create a SqueezeNet
   - `inchannels`: number of input channels.
   - `nclasses`: the number of output classes.
 
-See also [`Metalhead.squeezenet`](@ref).
+See also [`squeezenet`](@ref).
 """
 struct SqueezeNet
     layers::Any
diff --git a/src/utilities.jl b/src/utilities.jl
index 6f97e81d2..9f54107fd 100644
--- a/src/utilities.jl
+++ b/src/utilities.jl
@@ -15,8 +15,6 @@ end
 Convenience function for applying an activation function to the output after
 summing up the input arrays. Useful as the `connection` argument for the block
 function in [`resnet`](@ref).
-
-See also [`reluadd`](@ref).
 """
 addact(activation = relu, xs...) = activation(sum(xs))
 
@@ -26,8 +24,6 @@ addact(activation = relu, xs...) = activation(sum(xs))
 Convenience function for adding input arrays after applying an activation
 function to them. Useful as the `connection` argument for the block function in
 [`resnet`](@ref).
-
-See also [`addrelu`](@ref).
 """
 actadd(activation = relu, xs...) = sum(activation.(x) for x in xs)
 
diff --git a/toc.md b/toc.md
deleted file mode 100644
index c2599e837..000000000
--- a/toc.md
+++ /dev/null
@@ -1,11 +0,0 @@
-[Introduction](README.md)
-
-# Tutorials
-
-* [Quickstart](docs/tutorials/quickstart.md)
-
-# Developer guide
-
-* [Contributing](docs/dev-guide/contributing.md)
-
-[API Reference](docstrings.md)
\ No newline at end of file