diff --git a/docs/source/installation.mdx b/docs/source/installation.mdx
index caf22488f..ac2928cd2 100644
--- a/docs/source/installation.mdx
+++ b/docs/source/installation.mdx
@@ -91,6 +91,32 @@ Big thanks to [wkpark](https://github.com/wkpark), [Jamezo97](https://github.com
 </hfoption>
 </hfoptions>
 
+## Multi-backend preview release (+ compilation)
+
+Please follow these steps to install bitsandbytes with device-specific backend support other than CUDA:
+
+<hfoptions id="Alternative backend">
+<hfoption id="AMD ROCm">
+
+For a ROCm specific install:
+
+```bash
+git clone --depth 1 https://github.com/TimDettmers/bitsandbytes.git && cd bitsandbytes/
+git checkout multi-backend-refactor
+
+# Install dependencies
+apt install hipblaslt
+pip install -r requirements-dev.txt
+
+# Compile & install
+cmake -DCOMPUTE_BACKEND=hip -S . #Use -DBNB_ROCM_ARCH="gfx90a;gfx942" to target specific gpu arch
+make
+pip install .
+```
+
+<hfoption>
+</hfoptions>
+
 ## PyTorch CUDA versions
 
 Some bitsandbytes features may need a newer CUDA version than the one currently supported by PyTorch binaries from Conda and pip. In this case, you should follow these instructions to load a precompiled bitsandbytes binary.
diff --git a/docs/source/rocm_installation.mdx b/docs/source/rocm_installation.mdx
deleted file mode 100644
index 5d4381e7d..000000000
--- a/docs/source/rocm_installation.mdx
+++ /dev/null
@@ -1,46 +0,0 @@
-# ROCm Installation
-
-Please follow these steps to install bitsandbytes on ROCm.
-
-<hfoptions id="rocm_install">
-<hfoption id="latest">
-
-For latest installation:
-
-```bash
-git clone https://github.com/TimDettmers/bitsandbytes.git && cd bitsandbytes/
-pip install -r requirements-dev.txt
-cmake -DCOMPUTE_BACKEND=hip -S . #Use -DBNB_ROCM_ARCH="gfx90a;gfx942" to target specific gpu arch
-make
-pip install .
-```
-
-<hfoption>
-<hfoption id="version specific">
-
-For ROCm specific versions:
-
-Install Dependencies:
-
-```bash
-# hipblaslt installation needed only for rocm<6.0
-apt install hipblaslt
-pip install --upgrade pip
-pip install einops lion_pytorch accelerate
-pip install git+https://github.com/ROCm/transformers.git
-```
-
-Install bitsandbytes from [ROCm](https://github.com/ROCm/bitsandbytes) repo:
-
-```bash
-git clone --recurse https://github.com/ROCm/bitsandbytes
-cd bitsandbytes
-# Checkout branch as needed
-# for rocm 5.7 - rocm5.7_internal_testing
-# for rocm 6.x - rocm6.2_internal_testing
-git checkout <branch>
-make hip
-python setup.py install
-```
-<hfoption>
-</hfoptions>