# uv got to keep up: a new installer announced!

Astral, [Charlie Marsh](https://twitter.com/charliermarsh)'s company best known for Ruff, announced [uv](https://github.com/astral-sh/uv) today, "an extremely fast Python package installer and resolver, written in Rust." uv is intended as a "drop-in replacement for pip and [pip-compile](https://pip-tools.readthedocs.io/en/stable/cli/pip-compile/)."

Some early feedback looks very encouraging, e.g.,

<blockquote class="twitter-tweet" data-conversation="none" data-dnt="true"><p lang="en" dir="ltr">It looks like uv will bring dependency install times for <a href="https://twitter.com/dagster?ref_src=twsrc%5Etfw">@dagster</a> down from 1m26s to 18s, and a blistering 29ms with cache!<a href="https://t.co/z7MXOFKbiw">https://t.co/z7MXOFKbiw</a><br><br>Amazing work <a href="https://twitter.com/charliermarsh?ref_src=twsrc%5Etfw">@charliermarsh</a> and team!</p>&mdash; colton (@coltonpadden) <a href="https://twitter.com/coltonpadden/status/1758239894881661327?ref_src=twsrc%5Etfw">February 15, 2024</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

<blockquote class="twitter-tweet" data-dnt="true"><p lang="en" dir="ltr">I just swapped a Docker image from &quot;pip install --no-cache-dir -r requirements.txt&quot; to &quot;uv pip install --no-cache -r requirements.txt&quot; and the build time went from 67.8s to 15.4s. Absolutely insane. <a href="https://t.co/VwOIfIpnQP">https://t.co/VwOIfIpnQP</a></p>&mdash; Senior PowerPoint Engineer (@ryxcommar) <a href="https://twitter.com/ryxcommar/status/1758226228568277099?ref_src=twsrc%5Etfw">February 15, 2024</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

uv is not yet at full feature parity with pip and pip-compile, but given Astral's track record, we can expect that to change quickly.

An exciting announcement that came in parallel is that Astral is [taking stewardship of rye](https://lucumr.pocoo.org/2024/2/15/rye-grows-with-uv/), the experimental all-in-one tool for package and project management also written in Rust. The hope is that uv and rye will ultimately develop into a single tool that will install Python, install packages, build packages, create and manage virtual environments, manage dependencies, etc.

I was working on the rye entry for the Python Developer Tooling Handbook today; I've already got some revising to do!

Update: [@ryxcommar shared instructions and explanation for using uv in a Python Docker project.](https://ryxcommar.com/2024/02/15/how-to-cut-your-python-docker-builds-in-half-with-uv/)
