gguf-merge-tool 10.0

Professional GUI + CLI tool for downloading, verifying and merging sharded GGUF models

GGUF Merge Tool v10.0 https://badge.fury.io/py/gguf-merge-tool.svg https://img.shields.io/badge/python-3.8+-blue.svg https://img.shields.io/badge/License-MIT-yellow.svg

Professional GUI + CLI tool for downloading, verifying and merging sharded GGUF models from Hugging Face.

✨ What's New in Version 10.0 Feature Description Fixed Status Bars Progress displays smoothly and correctly during hash verification and merging operations Local File Loading "Local" button allows you to select a folder with GGUF files on disk and work with them without downloading Repository History Dropdown list in the repository input field — saves the last 10 queries 🚀 Key Features Download GGUF files from Hugging Face with resume support.

Integrity verification (SHA-256) of downloaded files.

Merge parts (splits) into a single GGUF file using llama-gguf-split.

Group files by quantization and base name.

Multithreaded downloading (auto-detects number of cores).

Real ETA and speed (MB/s) during download.

CLI mode for automation.

Persistent token and llama.cpp path across sessions.

📦 Installation From PyPI (recommended) bash pip install gguf-merge-tool After installation, two commands are available:

gguf-merge-gui — launch the graphical interface

gguf-merge — launch CLI mode

From Source bash git clone https://github.com/Andre17111978/gguf-merge-tool.git cd gguf-merge-tool pip install -r requirements.txt python gguf_merge_tool.py 🖥️ Usage (GUI) bash gguf-merge-gui

or

python gguf_merge_tool.py Step-by-step Interface: Field Description Repository Enter the Hugging Face repository ID (e.g., unsloth/Qwen3.5-122B-A10B-GGUF). History automatically saves the last 10 queries. Branch Usually main. Token Optional, for private repositories (format: hf_xxxxxxxxxxxxxxxxxx). Download Folder Where to save files. llama.cpp Path Specify the folder containing llama-gguf-split (required for merging). Files List of GGUF files with checkboxes, automatically grouped by quantization. Control Buttons: Button Function Load Fetch the list of files from the repository. Local Select a folder with GGUF files on disk. Select All / Download All Mass operations. Download Download selected files. Verify Check SHA-256 hashes. Merge Combine selected parts. ⌨️ Usage (CLI) Download Specific Files bash gguf-merge download -r unsloth/Qwen3.5-122B-A10B-GGUF -d ./models -f model-00001-of-00003.gguf model-00002-of-00003.gguf Download Entire Repository bash gguf-merge download -r unsloth/Qwen3.5-122B-A10B-GGUF -d ./models Download with Token (Private Repository) bash gguf-merge download -r your-username/your-private-model -d ./models -t hf_xxxxxxxxxxxxxxxxxx Merge Parts bash gguf-merge merge -d ./models -o ./models/merged_model.gguf -f CLI Arguments Command download:

Argument Description -d, --dir Save folder (required) -r, --repo Repository ID (required) -b, --branch Branch (default: main) -t, --token Hugging Face token -f, --files List of files to download Command merge:

Argument Description -d, --dir Folder with parts (required) -o, --output Output file path -f, --force Overwrite existing file ⚙️ Requirements Python >= 3.8

Dependencies (installed automatically): huggingface_hub >= 0.20.0

tqdm >= 4.60.0

requests >= 2.28.0

gguf >= 0.1.0 (optional, for reading metadata)

Additional Tool Required: llama-gguf-split (download from llama.cpp releases)

🛠️ Troubleshooting llama-gguf-split Not Found Download from llama.cpp releases

Place it in the tools/ folder or specify the path in GUI settings

HTTP 401/403 (Unauthorized/Forbidden) Enter a valid Hugging Face token

Ensure the token has access to the repository

Insufficient Disk Space Free up space or change the download folder

You can modify DISK_SPACE_MARGIN in the code (default: 1.1 = 10% buffer)

GUI Doesn't Start on Linux bash sudo apt-get install python3-tk # Debian/Ubuntu brew install python-tk # macOS 📝 License Distributed under the MIT License. See the LICENSE file for details.

🙏 Acknowledgments llama.cpp for the merge utility

Hugging Face for hosting models

The community for testing and ideas

⭐ Support the Project If you find this tool useful, please star it on GitHub!

Made with ❤️ for the AI community