Skip to main content

Unified GGUF wrapper for llama.cpp and Ollama

Project description

ggufy: Unified GGUF Model Wrapper

CI


Support This Project


ggufy unifies your local GGUF models from llama.cpp and Ollama into a single, easy-to-use collection. It handles symlinking, smart execution (local vs cloud), and provides a cleaner CLI experience.

Features

  • Unified Discovery: Finds models in ~/Library/Caches/llama.cpp and ~/.ollama/models/blobs.
  • Smart Symlinking: Creates a consistent ~/.guffy/models directory (or custom path).
  • Intelligent Runner: Automatically chooses between local GGUF and Ollama Cloud.
  • Audio/TTS Support: Detects audio models by name (containing "audio") and routes them to llama-cli.
  • Clean UX: Non-blocking lists, quiet output, and consistent flags.

Installation

Install via Cargo (Rust) or pip (Python, installs pre-built Rust binaries):

# From crates.io
cargo install ggufy

# From PyPI (downloads Rust binary via maturin)
pip install ggufy

# From source
git clone https://github.com/nbiish/ggufy.git
cd ggufy
cargo build --release
# Binaries will be in target/release/

After installation, ensure the binaries are on your PATH:

# For cargo installations
export PATH="$HOME/.cargo/bin:$PATH"

# For pip installations (usually automatic)
which ggufy

Quick Start

  1. List & Link: ggufy list (Refreshes links in ~/.guffy/models)
  2. Run Smart: ggufy run <model> (Uses local if available, else cloud)
  3. Simple Prompt: ggufy simple <model> "prompt"

Commands

Command Description
ggufy list Lists all local models and refreshes symlinks.
ggufy link Refreshes symlinks without listing.
ggufy locate <regex> Finds absolute paths of models matching pattern.
ggufy run <model> [args] Runs model via llama-server or ollama run.
ggufy simple <model> "txt" One-shot prompt. Quiet mode for local, chat mode for cloud.
ggufy llama <cmd> Wrappers for llama-server and llama-cli.
ggufy ollama <cmd> Wrappers for ollama serve/run.
ggufy -o <args> Passthrough to ollama <args>.
ggufy -c <args> Passthrough to llama-* tools (e.g. ggufy -c server ...).

Configuration

  • Link Directory: export GGUFY_MODELS_DIR="$HOME/.ggufy" (or use --link-dir)
  • Cache Override: export LLAMA_CPP_CACHE_DIR="..."
  • Ports: Default 12434 (llama), 11434 (ollama). Override with --port.

Audio/Multimodal

  • Audio Models: If the model name contains "audio", ggufy automatically uses llama-cli for execution.
  • Multimodal: Pass projection files manually: ggufy run model -- --mmproj path/to/mmproj.

Directory Structure

.
├── .github
│   └── workflows
├── docs
│   └── llama_cpp_overview.md
├── src
│   ├── bin
│   └── main.rs
├── tests
│   ├── integration.rs
│   └── run.sh
├── Cargo.toml
├── pyproject.toml
├── README.md
└── LICENSE

License & Citation

See LICENSE and CONTRIBUTING.md.

Policies

@misc{ggufy2025,
  author/creator/steward = {ᓂᐲᔥ ᐙᐸᓂᒥᑮ-ᑭᓇᐙᐸᑭᓯ (Nbiish Waabanimikii-Kinawaabakizi), also known legally as JUSTIN PAUL KENWABIKISE, professionally documented as Nbiish-Justin Paul Kenwabikise, Anishinaabek Dodem (Anishinaabe Clan): Animikii (Thunder), descendant of Chief ᑭᓇᐙᐸᑭᓯ (Kinwaabakizi) of the Beaver Island Band and enrolled member of the sovereign Grand Traverse Band of Ottawa and Chippewa Indians},
  title/description = {ggufy},
  type_of_work = {Indigenous digital creation/software incorporating traditional knowledge and cultural expressions},
  year = {2025},
  publisher/source/event = {GitHub repository under tribal sovereignty protections},
  howpublished = {\url{https://github.com/nbiish/ggufy}},
  note = {Authored and stewarded by ᓂᐲᔥ ᐙᐸᓂᒥᑮ-ᑭᓇᐙᐸᑭᓯ (Nbiish Waabanimikii-Kinawaabakizi), also known legally as JUSTIN PAUL KENWABIKISE, professionally documented as Nbiish-Justin Paul Kenwabikise, Anishinaabek Dodem (Anishinaabe Clan): Animikii (Thunder), descendant of Chief ᑭᓇᐙᐸᑭᓯ (Kinwaabakizi) of the Beaver Island Band and enrolled member of the sovereign Grand Traverse Band of Ottawa and Chippewa Indians. This work embodies Indigenous intellectual property, traditional knowledge systems (TK), traditional cultural expressions (TCEs), and associated data protected under tribal law, federal Indian law, treaty rights, Indigenous Data Sovereignty principles, and international indigenous rights frameworks including UNDRIP. All usage, benefit-sharing, and data governance are governed by the COMPREHENSIVE RESTRICTED USE LICENSE FOR INDIGENOUS CREATIONS WITH TRIBAL SOVEREIGNTY, DATA SOVEREIGNTY, AND WEALTH RECLAMATION PROTECTIONS.}
}

Copyright

Copyright © 2025 ᓂᐲᔥ ᐙᐸᓂᒥᑮ-ᑭᓇᐙᐸᑭᓯ (Nbiish Waabanimikii-Kinawaabakizi), also known legally as JUSTIN PAUL KENWABIKISE, professionally documented as Nbiish-Justin Paul Kenwabikise, Anishinaabek Dodem (Anishinaabe Clan): Animikii (Thunder), a descendant of Chief ᑭᓇᐙᐸᑭᓯ (Kinwaabakizi) of the Beaver Island Band, and an enrolled member of the sovereign Grand Traverse Band of Ottawa and Chippewa Indians. This work embodies Traditional Knowledge and Traditional Cultural Expressions. All rights reserved.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

ggufy-0.2.0.tar.gz (144.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

ggufy-0.2.0-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.7 MB view details)

Uploaded Python 3manylinux: glibc 2.17+ x86-64

File details

Details for the file ggufy-0.2.0.tar.gz.

File metadata

  • Download URL: ggufy-0.2.0.tar.gz
  • Upload date:
  • Size: 144.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: maturin/1.10.2

File hashes

Hashes for ggufy-0.2.0.tar.gz
Algorithm Hash digest
SHA256 31f0ddabd9d75a8de5f7aab81a57f9e1296c24c5eb421340da36f1b5b61693f5
MD5 02e1a4da3fadca2787353f9559eff483
BLAKE2b-256 f0b9563250f3be675740c1657f5af38502ae8a8277d8980bdc462095447f45ec

See more details on using hashes here.

File details

Details for the file ggufy-0.2.0-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.

File metadata

File hashes

Hashes for ggufy-0.2.0-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 d31398967257a991cfee6a97d3ce470ba1e62dad6e27e7579a4ded02f6696f3d
MD5 62c3a1c33f3041ac6305cf105d3689be
BLAKE2b-256 26a02f615b1286116e96aeb4277443df57887b9e2ae4156d1d89eefa2a209cee

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page