https://github.com/priyansusahoo/ollama-webui
Streamlined Ollama WebUI Setup: Automated Scripts, LLM Integration, and Desktop Shortcut Creation
https://github.com/priyansusahoo/ollama-webui
aifordevelopers aiprivacy aisetupscripts automation desktop-entry developertools efficientai linuxai llm localai localllm machinelearning offlineai ollama ollama-webui opensource privacyfocusedai selfhostedai ubuntu
Last synced: 16 days ago
JSON representation
Streamlined Ollama WebUI Setup: Automated Scripts, LLM Integration, and Desktop Shortcut Creation
- Host: GitHub
- URL: https://github.com/priyansusahoo/ollama-webui
- Owner: Priyansusahoo
- License: mit
- Created: 2025-01-25T17:20:38.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-01-26T06:52:19.000Z (3 months ago)
- Last Synced: 2025-04-14T19:07:31.104Z (16 days ago)
- Topics: aifordevelopers, aiprivacy, aisetupscripts, automation, desktop-entry, developertools, efficientai, linuxai, llm, localai, localllm, machinelearning, offlineai, ollama, ollama-webui, opensource, privacyfocusedai, selfhostedai, ubuntu
- Language: Shell
- Homepage:
- Size: 46.9 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Funding: .github/FUNDING.yml
- License: LICENSE
Awesome Lists containing this project
README
# About:
- Automated scripts to simplify the process of Ollama and its WebUI Launch in one click on Ubuntu 24.04.1 LTS (Gnome).
- Setting up a Large Language Model (LLM) like `DeepSeek-r1` with 8 billion parameters, and
- Creating a `desktop entry` for easy access to the WebUI (Auto start ollama server and Validations)
# SUPPORT:
⭐ **Star this repo if you find it useful!** ⭐
[](https://github.com/sponsors/Priyansusahoo)
# INSTALLATION:
## Ollama & Ollama-WebUI Installation and Creating Desktop Entry:
### UBUNTU 24.04.1 LTS (Gnome) (TESTED)
## 1. Install Ollama:
$ curl -fsSL https://ollama.com/install.sh | sh
$ sudo systemctl enable ollama
$ sudo systemctl start ollama
## 2. Install an LLM of your choice
### Select an LLM of your choice from [here](https://ollama.com/search).
- For demonstration I will install DeepSeek-r1 8 billion parameters (4.9 GB):
$ ollama run deepseek-r1:8b
- This will take some time, after successfully installing
>>> /bye
## 3. [Install Ollama-WebUI](https://snapcraft.io/install/ollama-webui/ubuntu):
- prerequisites (SNAP):
$ sudo apt update
$ sudo apt install snapd
- Ollama-WebUI:
$ sudo snap install ollama-webui --beta
## 4. Creating a Desktop Entry for Ollama-WebUI to access it from the app drawer
- Save this `ollama-launcher.sh` to ~/.scripts folder as mentioned `$HOME/scripts/ollama-launcher.sh`
$ cd
$ mkdir -p ~/.scripts/
$ cd scripts/
$ touch ollama-launcher.sh
$ vim ollama-launcher.sh
- `ollama-launcher.sh` :
```bash
#!/bin/bash
# Configuration
OLLAMA_HOST="127.0.0.1:11434"
OLLAMA_WEBUI_PORT=8080 # Changed to match your actual web UI port
# Check for required commands
check_commands() {
if ! command -v ollama &> /dev/null; then
echo "Error: Ollama is not installed. Please install Ollama first."
exit 1
fi
if ! command -v ollama-webui &> /dev/null; then
echo "Error: ollama-webui command not found. Please install it first."
exit 1
fi
}
# Function to check/start Ollama
start_ollama() {
if curl -s $OLLAMA_HOST >/dev/null; then
echo "✓ Ollama is already running"
return 0
fi
echo "⚠️ Ollama is not running. Starting now..."
# Try systemd service first
if command -v systemctl >/dev/null && systemctl start ollama 2>/dev/null; then
echo "✓ Ollama started via system service"
else
echo "✓ Starting Ollama in background..."
nohup ollama serve > /dev/null 2>&1 &
fi
# Wait for Ollama to become available
echo -n "⏳ Waiting for Ollama to start"
local max_attempts=15
for ((i=1; i<=$max_attempts; i++)); do
if curl -s $OLLAMA_HOST >/dev/null; then
echo -e "\n✓ Ollama is ready!"
return 0
fi
echo -n "."
sleep 1
done
echo -e "\n❌ Failed to start Ollama after $max_attempts attempts"
exit 1
}
# Function to handle port conflicts
handle_port_conflict() {
echo "🛑 Port $OLLAMA_WEBUI_PORT is already in use"
# Find PID using the port
pid=$(ss -tulpn | grep ":$OLLAMA_WEBUI_PORT" | awk '{print $7}' | cut -d= -f2 | cut -d, -f1)
if [ -n "$pid" ]; then
echo "🔍 Found existing process (PID: $pid)"
read -p "❓ Do you want to kill this process and continue? [y/N] " response
if [[ "$response" =~ ^[Yy]$ ]]; then
kill -9 $pid
echo "✅ Killed process $pid"
return 0
else
echo "❎ Exiting script"
exit 1
fi
else
echo "⚠️ Port in use but no process found - check your system"
exit 1
fi
}
# Function to start web UI
start_webui() {
if ss -tuln | grep -q ":$OLLAMA_WEBUI_PORT"; then
echo "✓ Web UI already running"
return 0
fi
echo "🚀 Starting Ollama Web UI..."
if ! nohup ollama-webui --port $OLLAMA_WEBUI_PORT > /dev/null 2>&1 & then
handle_port_conflict
nohup ollama-webui --port $OLLAMA_WEBUI_PORT > /dev/null 2>&1 &
fi
echo -n "⏳ Waiting for Web UI"
local max_attempts=10
for ((i=1; i<=$max_attempts; i++)); do
if ss -tuln | grep -q ":$OLLAMA_WEBUI_PORT"; then
echo -e "\n✓ Web UI is ready!"
return 0
fi
echo -n "."
sleep 1
done
echo -e "\n❌ Failed to start Web UI after $max_attempts attempts"
exit 1
}
# Open browser function
open_browser() {
echo "🌐 Opening browser at http://localhost:$OLLAMA_WEBUI_PORT"
xdg-open "http://localhost:$OLLAMA_WEBUI_PORT"
}
# Main script execution
check_commands
start_ollama
start_webui
open_browser
```
- After copying the content to `ollama-launcher.sh`, give the required permissions
$ chmod +x ollama-launcher.sh
- Copy `ollama-WebUI-Create-desktop-entry.sh` and save it in ~/.scripts folder which we previously created in the User's HOME directory and give the required permissions
$ cd
$ cd scripts/
$ touch ollama-WebUI-Create-desktop-entry.sh
$ vim ollama-WebUI-Create-desktop-entry.sh
- `ollama-WebUI-Create-desktop-entry.sh` :
```bash
#!/bin/bash
# Create desktop entry
cat > ~/.local/share/applications/ollama-webui.desktop <