Skip to content
OpenCharly

llama-cpp

Recipe card from the charly-jupyter plugin (Images — the deployable catalog).

llama-cpp – llama.cpp prebuilt binaries and GGUF tools

Section titled “llama-cpp – llama.cpp prebuilt binaries and GGUF tools”

Candy Properties

Section titled “Candy Properties”
Property Value
Dependencies None
Ports —
Service —
Install files charly.yml, task:

What It Installs

Section titled “What It Installs”

Downloads the latest llama.cpp release from GitHub into ~/llama.cpp:

Binaries: llama-quantize, llama-cli, shared libraries (lib*.so*)

Python tools: convert_hf_to_gguf.py, gguf-py package (from source tarball)

Environment

Section titled “Environment”
Variable Value Purpose
LLAMA_CPP_PATH ~/llama.cpp Location of llama.cpp binaries
PATH (appended) ~/llama.cpp Makes llama-quantize/llama-cli available

Design: Tier 1 “Post-install” Layer

Section titled “Design: Tier 1 “Post-install” Layer”

This candy has no pixi.toml and no depends. It downloads prebuilt binaries and sets environment variables. It is designed to be composed into environment-owning candies (Tier 2) via the candy: field.

The user-phase tasks run after the pixi environment is established by the parent candy. The gguf Python package (for programmatic GGUF access) is declared in the parent candy’s pixi.toml, not here.

Used In Candies

Section titled “Used In Candies”
Section titled “Related Candies”

Used In Boxes

Section titled “Used In Boxes”

When to Use This Skill

Section titled “When to Use This Skill”

Use when the user asks about:

  • llama.cpp binaries inside containers
  • GGUF model conversion (convert_hf_to_gguf.py)
  • llama-quantize or llama-cli
  • The LLAMA_CPP_PATH environment variable
Section titled “Related”
  • /charly-image:layer — candy authoring reference (charly.yml schema, task verbs, service declarations)
  • /charly-check:check — declarative testing (check: block, charly check box, charly check live)