133 lines
2.6 KiB
Markdown
133 lines
2.6 KiB
Markdown
# Ollama my big llama
|
|
### Working config 2026-05-13
|
|
|
|
sudo apt update && sudo apt upgrade -y
|
|
|
|
sudo apt install -y \
|
|
git curl wget htop btop tmux nano \
|
|
build-essential cmake python3 python3-pip \
|
|
bashtop
|
|
|
|
# Setup Nas disk per ollama model del dio cane
|
|
sudo mkdir -p /mnt/nas/ai
|
|
|
|
sudo apt install cifs-utils
|
|
sudo nano /etc/cifs-credentials
|
|
|
|
username=adminrack
|
|
password=TUO_PASSWORD
|
|
|
|
sudo nano /etc/fstab
|
|
|
|
# nas ai model location
|
|
//192.168.1.7/AI /mnt/nas/ai cifs credentials=/etc/cifs-credentials,vers=3.0,uid=ollama,gid=ollama,iocharset=utf8,file_mode=0775,dir_mode=0775,_netdev,nofail,x-systemd.automount 0 0
|
|
|
|
sudo systemctl daemon-reload
|
|
sudo mount -a
|
|
|
|
## organizzazione dei modelli
|
|
es DeepSeek-Coder-V2-Lite-Instruct-Q4_K_M.gguf
|
|
|
|
cd /mnt/nas/ai/Models
|
|
mkdir deepseek-q4
|
|
|
|
mv DeepSeek-Coder-V2-Lite-Instruct-Q4_K_M.gguf deepseek-q4/model.gguf
|
|
|
|
nano deepseek-q4/Modelfile
|
|
FROM ./model.gguf
|
|
|
|
### per eseguirlo
|
|
ollama run deepseek-q4
|
|
|
|
# Ollama user setup
|
|
sudo adduser ollama
|
|
|
|
sudo mkdir -p /home/ollama/.ollama
|
|
sudo chown -R ollama:ollama /home/ollama/.ollama
|
|
|
|
## Se non viene richiesta l'inserimento della password
|
|
sudo passwd ollama
|
|
|
|
## per autenticarsi come ollama
|
|
sudo -u ollama -i
|
|
|
|
# Ollama setup
|
|
curl -fsSL https://ollama.com/install.sh | sh
|
|
|
|
sudo systemctl edit ollama.service
|
|
|
|
[Unit]
|
|
Description=Ollama Server
|
|
After=network-online.target local-fs.target
|
|
Wants=network-online.target
|
|
|
|
[Service]
|
|
Type=simple
|
|
User=ollama
|
|
Group=ollama
|
|
Environment="OLLAMA_HOST=0.0.0.0:11434"
|
|
Environment="OLLAMA_ORIGINS=*"
|
|
Restart=always
|
|
RestartSec=5
|
|
|
|
sudo systemctl daemon-reload
|
|
sudo systemctl restart ollama
|
|
|
|
## Setup model from file per registrare il modello
|
|
cd /mnt/nas/ai/Models/deepseek-q4
|
|
ollama create deepseek-q4 -f Modelfile
|
|
|
|
# Webui setup
|
|
|
|
curl -fsSL https://get.docker.com | sh
|
|
sudo usermod -aG docker $USER
|
|
|
|
docker run -d \
|
|
-p 3000:8080 \
|
|
-v open-webui:/app/backend/data \
|
|
--name open-webui \
|
|
--restart always \
|
|
ghcr.io/open-webui/open-webui:main
|
|
|
|
# Continue config
|
|
|
|
name: Smart Local Router
|
|
version: 1.0.0
|
|
schema: v1
|
|
|
|
models:
|
|
- name: DeepSeek Coder Q4
|
|
provider: ollama
|
|
model: deepseek-coder-v2-q4
|
|
apiBase: http://192.168.1.250:11434
|
|
roles:
|
|
- chat
|
|
- edit
|
|
- apply
|
|
|
|
- name: DeepSeek Coder Q8
|
|
provider: ollama
|
|
model: deepseek-coder-v2-q8
|
|
apiBase: http://192.168.1.250:11434
|
|
roles:
|
|
- chat
|
|
- edit
|
|
- apply
|
|
|
|
- name: Qwen Agent
|
|
provider: ollama
|
|
model: qwen2.5-coder:14b
|
|
apiBase: http://192.168.1.250:11434
|
|
roles:
|
|
- chat
|
|
- edit
|
|
- apply
|
|
|
|
context:
|
|
- provider: code
|
|
- provider: diff
|
|
- provider: terminal
|
|
- provider: problems
|
|
- provider: folder
|
|
- provider: repo-map
|
|
- provider: os |