██████╗ ███████╗███████╗ █████╗     ███████╗██████╗ ███████╗███████╗
██╔══██╗██╔════╝╚══███╔╝██╔══██╗    ╚════██║╚════██╗╚════██║╚════██║
██████╔╝█████╗    ███╔╝ ███████║        ██╔╝ █████╔╝    ██╔╝    ██╔╝
██╔══██╗██╔══╝   ███╔╝  ██╔══██║       ██╔╝ ██╔═══╝    ██╔╝    ██╔╝
██║  ██║███████╗███████╗██║  ██║       ██║  ███████╗   ██║     ██║
╚═╝  ╚═╝╚══════╝╚══════╝╚═╝  ╚═╝       ╚═╝  ╚══════╝   ╚═╝     ╚═╝

A full-featured, interactive Bash manager for Ollama LLMs

---

Overview

Ollama Manager is a single-file Bash script that gives you a polished, menu-driven TUI (terminal UI) for managing your local LLMs through Ollama. No Python, no Node — just bash, curl, jq, and zstd.

It lets you browse the full Ollama model library, download models with quantization control, run interactive chats, manage your model collection, and view detailed hardware/GPU status — all from one unified interface.

---

Features

Category	Capabilities
Download	Browse 200+ models from ollamadb.dev, search & filter by type, paginated table view, queue multiple downloads
Quantization	Choose from q4_0, q4_K_M, q5_0, q5_K_M, q8_0, f16, or default
Chat	Interactive terminal chat, system prompt presets, custom temperature & context size, chat history saving
Model Management	View, delete, copy/rename, update all, create custom Modelfiles
API Mode	Load any model into the REST API server and get ready-to-use curl examples
System Info	RAM, CPU, GPU (NVIDIA/AMD), disk usage, Ollama version & update check
Smart Checks	RAM/disk warnings before download, resource usage estimates
Auto-setup	Detects and installs missing dependencies (curl, jq, zstd), installs Ollama if not present
Offline Fallback	Built-in curated model list when ollamadb.dev is unreachable
Caching	Model list cached for 1 hour to minimize API calls

System Prompt Presets

• None (default)
• Persian / English Translator
• Code Reviewer
• DevOps Assistant
• Bash Script Expert
• Security Analyst
• Data Scientist
• Technical Writer
• Custom (type your own)

---

Requirements

Dependency	Purpose	Auto-installed?
curl	API calls & Ollama install	Yes (apt/yum/pacman)
jq	JSON parsing	Yes (apt/yum/pacman)
zstd	Ollama extraction	Yes (apt/yum/pacman)
ollama	LLM runtime	Yes (prompted)

OS: Linux (Debian/Ubuntu, RHEL/CentOS/Fedora, Arch)

The script also works on WSL2 (Windows Subsystem for Linux).

---

Installation

# Clone the repository
git clone https://github.com/reza7277/ollama-script.git
cd ollama-script

# Make it executable
chmod +x ollama_manager.sh

# Run it
./ollama_manager.sh

Or as a one-liner:

bash <(curl -fsSL https://raw.githubusercontent.com/reza7277/ollama-script/refs/heads/main/ollama_manager.sh)

---

Usage

./ollama_manager.sh

On first launch the script will:

1. Check for curl, jq, and zstd — offer to install any that are missing
2. Check if Ollama is installed — offer to install it automatically
3. Check if the Ollama service is running — offer to start it

Main Menu

╔══════════════════════════════════════════════════════════╗
║                       MAIN MENU                          ║
╠══════════════════════════════════════════════════════════╣
║  1   Download & Setup Models                             ║
║      Browse & download from the full Ollama library      ║
║                                                          ║
║  2   Run Downloaded Models (Offline)                     ║
║      Chat with your installed models — no internet needed║
║                                                          ║
║  3   Model Management                                    ║
║      View, delete, copy, update models                   ║
║                                                          ║
║  4   System Info & Status                                ║
║      Hardware info, GPU, Ollama logs, update check       ║
║                                                          ║
║  5   Exit                                                ║
╚══════════════════════════════════════════════════════════╝

Download Section Commands

Command	Action
n	Next page
p	Previous page
s	Search by name or description
i <num>	Show detailed info for model
d <num>	Download model immediately
a <num> [num ...]	Add to download queue
q	Download queued models & exit
b	Back to main menu

Configuration

Settings are stored in ~/.ollama_manager/config.conf:

DEFAULT_TEMPERATURE=0.7
DEFAULT_CONTEXT=4096
AUTO_START_OLLAMA=true
SAVE_CHAT_HISTORY=true
MODELS_PER_PAGE=15

Chat History

When SAVE_CHAT_HISTORY=true, every chat session is saved to:

~/.ollama_manager/chats/YYYY-MM-DD_HH-MM-SS_modelname.txt

You can view saved sessions from System Info > View saved chat history.

API Server Mode

Select Run > Start API Server only to load a model and get instant copy-paste API examples:

# Generate
curl http://localhost:11434/api/generate \
  -d '{"model":"llama3.2","prompt":"Hello!"}'

# Chat
curl http://localhost:11434/api/chat \
  -d '{"model":"llama3.2","messages":[{"role":"user","content":"Hello!"}]}'

---

Model Categories

Filter	Description
All	Every available model
Chat	General-purpose conversational models
Vision	Multimodal models (image + text)
Code	Code generation & review specialists
Embedding	Vector embedding models for RAG/search

---

Popular Models (Built-in Fallback List)

Model	Size	Best For
llama3.2	2.0 GB	General chat, fast
llama3.1	4.7 GB	Balanced quality
mistral	4.1 GB	Fast & accurate
qwen2.5	4.7 GB	Multilingual
deepseek-r1	4.7 GB	Reasoning
qwen2.5-coder	4.7 GB	Code
gemma3	3.3 GB	Multimodal
nomic-embed-text	274 MB	Embeddings / RAG

---

File Structure

~/.ollama_manager/
├── config.conf          # User settings
├── model_cache.json     # Cached model list (1h TTL)
├── ollama.log           # Ollama service log
└── chats/               # Saved chat sessions
    └── 2026-02-19_...   # Timestamped transcripts

---

Screenshots

Coming soon — contributions welcome!

---

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you'd like to change.

1. Fork the repo
2. Create your branch (git checkout -b feature/my-feature)
3. Commit your changes (git commit -m 'Add my feature')
4. Push to the branch (git push origin feature/my-feature)
5. Open a Pull Request

---

License

This project is licensed under the MIT License — see the LICENSE file for details.

---

Author

Reza — @Web3loverz

Join the Telegram channel for updates, tips, and community support.

---

_{Made with ❤️ for the self-hosted AI community}