llmfit 0.9.4

Unofficial PyPI distribution of llmfit – the LLM model management CLI

llmfit via PyPI

Unofficial PyPI distribution of llmfit — the LLM model management CLI.

This package downloads pre-built binaries from the upstream GitHub releases and repackages them as Python wheels so you can install llmfit with pip or uv without a Rust toolchain.

Installation

pip install llmfit
# or
uv add llmfit

After installation the llmfit command is available on your PATH.

llmfit --help

You can also use this package to install and update llmfit via uv tool:

uv tool install -U llmfit

Supported platforms

See Rust platform support for more information. Refer to the the upstream llmfit project for authoritative requirements.

Platform	Architecture	Requirements
Linux (glibc)	x86_64	kernel ≥ 3.2, glibc ≥ 2.17
Linux (glibc)	aarch64	kernel ≥ 4.1, glibc ≥ 2.17
Linux (musl)	x86_64	musl ≥ 1.2.5
Linux (musl)	aarch64	musl ≥ 1.2.5
macOS	x86_64 (Intel)	macOS ≥ 10.12
macOS	arm64 (Apple Silicon)	macOS ≥ 11.0
Windows	x86_64	Windows 10+ or Windows Server 2016+
Windows	ARM64

Version correspondence

The version of this package always matches the upstream llmfit release tag (with the leading v stripped). llmfit==0.8.6 contains v0.8.6 of the upstream binary.

About this package

This is an unofficial redistribution. The llmfit binary is the work of Alex Jones and contributors, released under the MIT License. See LICENSE for details.

Source for this packaging wrapper: https://github.com/JEHoctor/llmfit-pypi

---

For maintainers of this repository

How it works

1. A nightly GitHub Actions workflow (check_upstream.yml) compares the latest upstream tag with the published PyPI version.
2. If they differ, it triggers build_and_publish.yml with the new tag.
3. build_and_publish.yml calls build_wheels.py, which downloads each platform archive from GitHub Releases, verifies its SHA256 checksum, extracts the binary, and constructs a platform-tagged wheel for the llmfit package.
4. All wheels are published to PyPI via OIDC Trusted Publisher (no API tokens stored in repository secrets).

You can also trigger a build manually from the Actions tab, providing the version tag (e.g. v0.8.6).