From bb63309d1cb43baff7c4730feae1ce112a05e801 Mon Sep 17 00:00:00 2001
From: April M <36110273+aimurphy@users.noreply.github.com>
Date: Tue, 13 Jan 2026 09:18:08 -0800
Subject: [PATCH] issue 762 and 765

---
 docs/docs/_partial-prereq-common.mdx |  4 +++-
 docs/docs/get-started/docker.mdx     | 33 ++++++++++++++++++----------
 docs/docs/support/troubleshoot.mdx   | 27 +++++++++++++++++++++++
 3 files changed, 52 insertions(+), 12 deletions(-)

diff --git a/docs/docs/_partial-prereq-common.mdx b/docs/docs/_partial-prereq-common.mdx
index 4682d8cc..621273c7 100644
--- a/docs/docs/_partial-prereq-common.mdx
+++ b/docs/docs/_partial-prereq-common.mdx
@@ -15,4 +15,6 @@ If a provider offers only one type, you must select two providers.
       <PartialOllamaModels />
       :::
 
-* Optional: Install GPU support with an NVIDIA GPU, [CUDA](https://docs.nvidia.com/cuda/) support, and compatible NVIDIA drivers on the OpenRAG host machine. If you don't have GPU capabilities, OpenRAG provides an alternate CPU-only deployment.
\ No newline at end of file
+* Optional: Install GPU support with an NVIDIA GPU, [CUDA](https://docs.nvidia.com/cuda/) support, and compatible NVIDIA drivers on the OpenRAG host machine.
+If you don't have GPU capabilities, OpenRAG provides an alternate CPU-only deployment that is suitable for most use cases.
+The default CPU-only deployment doesn't prevent you from using GPU acceleration in external services, such as Ollama servers.
\ No newline at end of file
diff --git a/docs/docs/get-started/docker.mdx b/docs/docs/get-started/docker.mdx
index 811d3388..2f12d353 100644
--- a/docs/docs/get-started/docker.mdx
+++ b/docs/docs/get-started/docker.mdx
@@ -116,17 +116,7 @@ The following variables are required or recommended:
 
 3. Deploy the OpenRAG containers locally using the appropriate Docker Compose configuration for your environment:
 
-   * **GPU-accelerated deployment**: If your host machine has an NVIDIA GPU with CUDA support and compatible NVIDIA drivers, use the base `docker-compose.yml` file with the `docker-compose.gpu.yml` override.
-
-      ```bash title="Docker"
-      docker compose -f docker-compose.yml -f docker-compose.gpu.yml up -d
-      ```
-
-      ```bash title="Podman"
-      podman compose -f docker-compose.yml -f docker-compose.gpu.yml up -d
-      ```
-
-   * **CPU-only deployment** (default): If your host machine doesn't have NVIDIA GPU support, use the base `docker-compose.yml` file.
+   * **CPU-only deployment** (default, recommended): If your host machine doesn't have NVIDIA GPU support, use the base `docker-compose.yml` file.
 
       ```bash title="Docker"
       docker compose up -d
@@ -136,6 +126,27 @@ The following variables are required or recommended:
       podman compose up -d
       ```
 
+   * **GPU-accelerated deployment**: If your host machine has an NVIDIA GPU with CUDA support and compatible NVIDIA drivers, use the base `docker-compose.yml` file with the `docker-compose.gpu.yml` override:
+
+   ```bash title="Docker"
+   docker compose -f docker-compose.yml -f docker-compose.gpu.yml up -d
+   ```
+
+   ```bash title="Podman"
+   podman compose -f docker-compose.yml -f docker-compose.gpu.yml up -d
+   ```
+
+   tip:::
+   GPU acceleration isn't required for most use cases.
+   OpenRAG's CPU-only deployment doesn't prevent you from using GPU acceleration in external services, such as Ollama servers.
+
+   GPU acceleration is required only for specific use cases, typically involving customization of the ingestion flows or ingestion logic.
+   For example, writing alternate ingest logic in OpenRAG that uses GPUs directly in the container, or customizing the ingestion flows to use Langflow's Docling component with GPU acceleration instead of the `docling serve` service.
+
+   If you are deploying OpenRAG on an Amazon EC2 instance with GPU acceleration enabled, you must edit the `docker-compose.gpu.yml` file before deploying the OpenRAG containers.
+   For more information, see [Some images failed to pull on Amazon EC2 with GPU support](/support/troubleshoot#some-images-failed-to-pull-on-amazon-ec2-with-gpu-support).
+   :::
+
 4. Wait for the OpenRAG containers to start, and then confirm that all containers are running:
 
    ```bash title="Docker"
diff --git a/docs/docs/support/troubleshoot.mdx b/docs/docs/support/troubleshoot.mdx
index f4d69e6e..fa1cdaa4 100644
--- a/docs/docs/support/troubleshoot.mdx
+++ b/docs/docs/support/troubleshoot.mdx
@@ -35,6 +35,33 @@ open "/Applications/Python VERSION/Install Certificates.command"
 
 Replace `VERSION` with your installed Python version, such as `3.13`.
 
+### Some images failed to pull on Amazon EC2 with GPU support {#some-images-failed-to-pull-on-amazon-ec2-with-gpu-support}
+
+When installing OpenRAG on an Amazon EC2 instance with GPU acceleration enabled, some Docker images might fail to pull.
+
+To resolve this issue, do the following:
+
+1. [Stop the OpenRAG containers](/manage-services#stop-and-start-containers).
+
+2. In your `~/.openrag` directory, edit the `docker-compose.gpu.yml` file as follows:
+
+   ```yaml
+   services:
+   openrag-backend:
+      environment:
+         - NVIDIA_DRIVER_CAPABILITIES=compute,utility
+         - NVIDIA_VISIBLE_DEVICES=all
+      deploy:
+         resources:
+         reservations:
+            devices:
+               - driver: nvidia
+               count: all
+               capabilities: [gpu]
+   ```
+
+3. [Restart the OpenRAG containers](/manage-services#stop-and-start-containers).
+
 ### Application onboarding gets stuck on Google Chrome
 
 If the OpenRAG onboarding process gets stuck when using Google Chrome, try clearing your browser's cache.