From 26cda893f042d12eb1c27ea743e285261778bca8 Mon Sep 17 00:00:00 2001
From: kat <kat@bluebirdV.lan>
Date: Fri, 6 Jun 2025 15:12:20 +0200
Subject: [PATCH 1/2] Add ollama service to docker compose

---
 README.md          | 33 ++++++++++++++++++++++++---------
 docker-compose.yml | 17 +++++++++++++----
 2 files changed, 37 insertions(+), 13 deletions(-)

diff --git a/README.md b/README.md
index d623e8b..1ed61c7 100644
--- a/README.md
+++ b/README.md
@@ -47,18 +47,24 @@ Choose models based on your system capabilities:
 | **Chat** | `phi3:mini` | ~2.3GB | 4GB | Low-resource systems |
 
 
+### Installation Options
+
+Choose your preferred installation method:
+
+### Option 1: Direct Installation
 
-### Prerequisites (Required for Both Installation Methods)
+**Prerequisite: Ollama (for local AI models)**
+
+Install Ollama
 
-**1. Install Ollama** (for local AI models):
 ```bash
 # macOS
 brew install ollama
 
 # Or download from https://ollama.com
 ```
+ Start Ollama and install required models
 
-**2. Start Ollama and install required models**:
 ```bash
 ollama serve
 
@@ -69,11 +75,7 @@ ollama pull nomic-embed-text
 ollama pull qwen3:14b
 ```
 
-### Installation Options
-
-Choose your preferred installation method:
 
-### Option 1: Direct Installation
 
 **Additional Prerequisites:**
 - Python 3.8+
@@ -106,7 +108,10 @@ Choose your preferred installation method:
 
 ### Option 2: Docker Installation
 
-**Additional Prerequisites:**
+With this option, you don't need to separately install Ollama, it will automatically
+get started by docker compose.
+
+**Prerequisites:**
 - Docker and Docker Compose
 
 **Installation Steps:**
@@ -122,7 +127,17 @@ Choose your preferred installation method:
    docker-compose up
    ```
 
-3. **Open your browser** to `http://localhost:8501`
+3. **Install models**
+
+   ```
+   # embedding model
+   docker exec -it ollama ollama pull nomic-embed-text
+   
+   # chat model
+   docker exec -it ollama ollama pull qwen3:14b
+   ```
+
+4. **Open your browser** to `http://localhost:8501`
 
 ## 📖 How to Use
 
diff --git a/docker-compose.yml b/docker-compose.yml
index 3bf846b..f3d1f63 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -11,10 +11,19 @@ services:
       - "8501:8501"
     environment:
       # Configure Ollama connection (both env vars for compatibility)
-      - OLLAMA_HOST=http://host.docker.internal:11434
-      - OLLAMA_BASE_URL=http://host.docker.internal:11434
+      - OLLAMA_HOST=ollama:11434
+      - OLLAMA_BASE_URL=http://ollama:11434
     restart: unless-stopped
     volumes:
       - ./data:/app/data  # For persistent data storage
-    extra_hosts:
-      - "host.docker.internal:host-gateway" 
\ No newline at end of file
+
+  ollama:
+    image: docker.io/ollama/ollama:latest
+    container_name: ollama
+    pull_policy: always
+    tty: true
+    restart: always
+    environment:
+      - OLLAMA_KEEP_ALIVE=24h
+      - OLLAMA_HOST=0.0.0.0
+      - OLLAMA_PORT=11434
\ No newline at end of file

From 9c9cc245ca303c22bb95cf18d2da1954a2131e58 Mon Sep 17 00:00:00 2001
From: krasch <dev@krasch.io>
Date: Fri, 6 Jun 2025 15:48:33 +0200
Subject: [PATCH 2/2] Add volume and GPU configuration for ollama service

---
 README.md              |  4 ++++
 docker-compose.gpu.yml | 10 ++++++++++
 docker-compose.yml     |  7 ++++++-
 docker.env.example     |  8 --------
 4 files changed, 20 insertions(+), 9 deletions(-)
 create mode 100644 docker-compose.gpu.yml
 delete mode 100644 docker.env.example

diff --git a/README.md b/README.md
index 1ed61c7..54623b9 100644
--- a/README.md
+++ b/README.md
@@ -124,7 +124,11 @@ get started by docker compose.
 
 2. **Start with Docker Compose**:
    ```bash
+   # if you don't have a GPU
    docker-compose up
+   
+   # if you have a GPU
+   docker compose -f docker-compose.yml -f docker-compose.gpu.yml up
    ```
 
 3. **Install models**
diff --git a/docker-compose.gpu.yml b/docker-compose.gpu.yml
new file mode 100644
index 0000000..7f5be24
--- /dev/null
+++ b/docker-compose.gpu.yml
@@ -0,0 +1,10 @@
+services:
+  ollama:
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: 1
+              capabilities:
+                - gpu
\ No newline at end of file
diff --git a/docker-compose.yml b/docker-compose.yml
index f3d1f63..48c8697 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -26,4 +26,9 @@ services:
     environment:
       - OLLAMA_KEEP_ALIVE=24h
       - OLLAMA_HOST=0.0.0.0
-      - OLLAMA_PORT=11434
\ No newline at end of file
+      - OLLAMA_PORT=11434
+    volumes:
+      - ollama:/root/.ollama
+
+volumes:
+  ollama: { }
diff --git a/docker.env.example b/docker.env.example
deleted file mode 100644
index 75f04ba..0000000
--- a/docker.env.example
+++ /dev/null
@@ -1,8 +0,0 @@
-# Ollama Configuration
-# Set the base URL for your Ollama instance
-OLLAMA_BASE_URL=http://localhost:11434
-
-# Examples for different setups:
-# Local Ollama on host: http://host.docker.internal:11434
-# Remote Ollama server: http://your-ollama-server:11434
-# Docker network Ollama: http://ollama-container:11434 
\ No newline at end of file