diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
new file mode 100644
index 0000000..d537e2a
--- /dev/null
+++ b/.github/workflows/release.yml
@@ -0,0 +1,57 @@
+name: Release
+
+on:
+  workflow_dispatch:
+
+env:
+  BUILD_TYPE: Release
+  
+jobs:
+  create_release:
+    name: Create release
+    runs-on: ubuntu-latest
+    outputs:
+      upload_url: ${{steps.create_release.outputs.upload_url}}
+    steps:
+      - name: Check out repository
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      - name: Create release
+        id: create_release
+        uses: actions/create-release@v1
+        env:
+          GITHUB_TOKEN: ${{secrets.GITHUB_TOKEN}}
+        with:
+          draft: true
+          tag_name: ${{github.ref}}
+          release_name: Release ${{github.ref}}
+          
+  build-windows:
+    name: Build Windows
+    needs: create_release
+    runs-on: windows-latest
+    steps:
+    - uses: actions/checkout@v3.3.0
+      with:
+        submodules: recursive
+    
+    - name: Build Plugin
+      working-directory: ${{github.workspace}}/build
+      run: |
+        cmake.exe -G "Visual Studio 17 2022" -A x64  -T ClangCL ..
+        cmake --build . --config=release -j4
+
+    - name: Add LV2 Archive
+      run: Compress-Archive -Path ${{github.workspace}}\build\neural_amp_modeler.lv2 -Destination neural_amp_modeler.lv2.zip
+      
+    - name: Upload Plugin Asset
+      uses: actions/upload-release-asset@v1
+      env:
+        GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      with:
+        upload_url: ${{ needs.create_release.outputs.upload_url }}
+        asset_path: ./neural_amp_modeler.lv2.zip
+        asset_name: neural_amp_modeler.lv2.zip
+        asset_content_type: application/zip
diff --git a/CREDITS.md b/CREDITS.md
index 46ebc2e..e8eec39 100644
--- a/CREDITS.md
+++ b/CREDITS.md
@@ -4,6 +4,8 @@ https://github.com/sdatkinson/NeuralAmpModelerCore
 
 https://gitlab.com/libeigen/eigen
 
+https://github.com/jatinchowdhury18/RTNeural
+
 https://github.com/lv2/lv2
 
 In addition, the CMake structure and LV2 plugin structure are based on code from https://github.com/Dougal-s/Aether.
diff --git a/README.md b/README.md
index 9477a04..678b22c 100644
--- a/README.md
+++ b/README.md
@@ -1,8 +1,8 @@
 # neural-amp-modeler-lv2
 
-Bare-bones implementation of [Neural Amp Modeler](https://github.com/sdatkinson/neural-amp-modeler) (NAM) models in an LV2 plugin.
+LV2 plugin for using neural network machine learning amp models.
 
-**There is no user interface**. Setting the model to use requires that your LV2 host supports atom:Path parameters. Reaper does as of v6.82. Carla and Ardour do. If your favorite LV2 host does not support atom:Path, let them know you want it.
+**There is no custom plugin user interface**. Setting the model to use requires that your LV2 host supports atom:Path parameters. Reaper does as of v6.82. Carla and Ardour do. If your favorite LV2 host does not support atom:Path, let them know you want it.
 If you are looking for a GUI version, @brummer10 [has one here](https://github.com/brummer10/neural-amp-modeler-ui) that works for Linux and Windows. You may also be interested in the the version shipped with the [MOD Desktop App](https://github.com/moddevices/mod-desktop-app), or my digital pedalboard app [Stompbox](https://github.com/mikeoliphant/StompboxUI).
 
 To get the intended behavior, **you must run your audio host at the same sample rate the model was trained at** (usually 48kHz) - no resampling is done by the plugin.
@@ -11,13 +11,15 @@ For amp-only models (the most typical), **you will need to run an impulse repons
 
 ### Models and Performance
 
+The plugin supports both [Neural Amp Modeler (NAM)](https://github.com/sdatkinson/neural-amp-modeler) models and [RTNeural keras json models](https://github.com/jatinchowdhury18/RTNeural) (like those used by [Aida-X](https://github.com/AidaDSP/AIDA-X)).
+
 The best source of models is [ToneHunt](https://tonehunt.org/).
 
-NAM models are generally quite expensive to run. This isn't (much of) an issue on modern PCs, but you may have trouble running on less powerful hardware.
+NAM WaveNet models are generally quite expensive to run. This isn't (much of) an issue on modern PCs, but you may have trouble running on less powerful hardware.
 
 A Raspberry Pi 4 running a 64bit OS can run "standard" NAM models with a bit of room to spare for a cabinet IR and some lightweight effects.
 
-If you are having trouble running a "standard" model, try looking for "feather" (the least expensive) models. You can find a list of ["feather"-tagged models on ToneHunt](https://tonehunt.org/?tags=feather-mdl). Note that tagging models is up to the submitter, so not all "feather" models are tagged as such - you should be able to find more if you dig around.
+If you are having trouble running a "standard" model, try looking for "feather", or even "nano" (the least expensive) models. You can find a list of ["feather"-tagged models on ToneHunt](https://tonehunt.org/models?tags%5B0%5D=feather-mdl). Note that tagging models is up to the submitter, so not all "feather" models are tagged as such - you should be able to find more if you dig around.
 
 ### Building