1
2
3
4
5
6
7
8
# Using Homebrew
brew install ollama

# Manual download
curl https://ollama.ai/download/ollama-darwin.zip --output ollama-darwin.zip
unzip ollama-darwin.zip
sudo mv ollama /usr/local/bin/

1
2
3
4
5
6
7
8
9
10
11
# Debian/Ubuntu
curl -fsSL https://ollama.ai/install.sh | sh

# Fedora/RHEL
curl -fsSL https://ollama.ai/install.sh | sh

# Manual installation
wget https://ollama.ai/download/ollama-linux-amd64
chmod +x ollama-linux-amd64
sudo mv ollama-linux-amd64 /usr/local/bin/ollama

1
2
3
4
5
# Download from official website
winget install ollama
# Or use Docker
docker pull ollama/ollama

1
2
3
4
5
6
# Start Ollama service
ollama serve

# Start in background
ollama serve &

1
2
3
4
5
6
7
8
# Pull a model
ollama pull llama2
ollama pull mistral

# Run a model interactively
ollama run llama2
ollama run mistral

1
2
3
4
5
6
7
8
9
10
# Pull a specific model version
ollama pull llama2:13b
ollama pull mistral:7b

# List available models
ollama list

# Show model details
ollama show llama2

1
2
3
4
5
6
# Remove a specific model
ollama rm llama2

# Remove all models
ollama rm --all

1
2
3
4
5
6
# Start interactive chat
ollama run llama2

# Exit chat
/bye

1
2
3
4
5
6
# Run with a prompt
ollama run llama2 "Explain quantum computing"

# Pipe input
echo "Write a Python script" | ollama run codellama

1
2
3
4
5
6
7
8
# Generate text
curl http://localhost:11434/api/generate \
  -d '{
    "model": "llama2",
    "prompt": "Tell me a story"
  }'

# Chat completion
curl http://localhost:11434/api/chat \
  -d '{
    "model": "mistral",
    "messages": [
      {"role": "user", "content": "Hello"}
    ]
  }'

1
2
3
4
5
6
# Set model directory
export OLLAMA_MODELS=/path/to/models

# Configure GPU usage
export OLLAMA_CUDA_VISIBLE_DEVICES=0,1

1
2
3
4
5
# Example Modelfile
FROM llama2
PARAMETER temperature 0.7
SYSTEM You are a helpful assistant

1
2
3
4
5
6
# Create a custom model
ollama create mymodel -f Modelfile

# Use custom model
ollama run mymodel

1
2
3
# Pull quantized models
ollama pull llama2:7b-q4_0

1
2
3
# View Ollama logs
journalctl -u ollama

1
2
3
4
5
from langchain.llms import Ollama

llm = Ollama(model="mistral")
result = llm("Write a poem")

1
2
3
# Use with other CLI tools
ollama run llama2 "Summarize this:" | pbcopy