diff --git a/docs/docs/concepts/fleets.md b/docs/docs/concepts/fleets.md
index 5b72804d86..52894d2d52 100644
--- a/docs/docs/concepts/fleets.md
+++ b/docs/docs/concepts/fleets.md
@@ -91,9 +91,17 @@ are both acceptable).
 
     !!! info "Requirements" 
         Hosts should be pre-installed with Docker.
-        Systems with NVIDIA GPUs should also be pre-installed with CUDA 12.1 and
-        [NVIDIA Container Toolkit :material-arrow-top-right-thin:{ .external }](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html).
-        The user should have `sudo` access.
+
+        === "NVIDIA"
+            Systems with NVIDIA GPUs should also be pre-installed with CUDA 12.1 and
+            [NVIDIA Container Toolkit :material-arrow-top-right-thin:{ .external }](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html).
+
+        === "AMD"
+            Systems with AMD GPUs should also be pre-installed with AMDGPU-DKMS kernel driver (e.g. via
+            [native package manager :material-arrow-top-right-thin:{ .external }](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/native-install/index.html)
+            or [AMDGPU installer :material-arrow-top-right-thin:{ .external }](https://rocm.docs.amd.com/projects/install-on-linux/en/latest/install/amdgpu-install.html).)
+
+        The user should have passwordless `sudo` access.
 
     ??? info "Environment variables"
         For SSH fleets, it's possible to pre-configure environment variables. 
diff --git a/examples/accelerators/amd/README.md b/examples/accelerators/amd/README.md
index 5eb24f41e9..874852e726 100644
--- a/examples/accelerators/amd/README.md
+++ b/examples/accelerators/amd/README.md
@@ -1,9 +1,7 @@
 # AMD
 
-Since [0.18.11 :material-arrow-top-right-thin:{ .external }](https://github.com/dstackai/dstack/releases/0.18.11rc1){:target="_blank"},
-you can specify an AMD GPU under `resources`. Below are a few examples.
-
-> AMD accelerators are currently supported only with the [`runpod`](https://dstack.ai/docs/reference/server/config.yml/#runpod) backend.
+If you're using the `runpod` backend or have set up an [SSH fleets](https://dstack.ai/docs/concepts/fleets#ssh-fleets) 
+with on-prem AMD GPUs, you can use AMD GPUs.
 
 ## Deployment