llama.cpp

Release: v1.0.0 (2026-11-13)Commit: b1b770dLicense: MIT

Runs GGUF language models on CPUs and consumer GPUs, with a local server.

inference

v1.0.0releasellama.cppC++MIT

FactsLicenseMITLanguageC++KindinferenceRelease2026-11-13PlatformsmacOS, Linux, WindowsLinks to the projectpermissive

$ brew install llama.cpp
# use with Freeweights
$ llama-server -m ./<file>.gguf --port 8080

Overview

Runs GGUF language models on CPUs and consumer GPUs, with a local server.

Topicsggml

Freeweights links to the project and does not host tool binaries.

$ brew install llama.cpp
# use with Freeweights
$ llama-server -m ./<file>.gguf --port 8080

$ brew install llama.cpp

Checked against the project's README on 2026-11-14. The project's own instructions win when they differ: install docs

Freeweights links to the project and does not host tool binaries.

Run a catalog model

$ llama-server -m ./<file>.gguf --port 8080

Download the weights from Freeweights first (HF_ENDPOINT or fw.py), then point the tool at the folder.

Freeweights links to the project and does not host tool binaries.

From:

ggml-org <github.com/ggml-org>Newsgroups: alt.freeweights.toolsSubject: llama.cppRelease: v1.0.0 (2026-11-13)Commit: b1b770dLicense: MIT

Freeweights links to the project and does not host tool binaries.

Get it

Works with

11 catalog models. See the list