pickfit

BrowseLocal AI tools

Berth

Schedules GPU memory so chat, image, voice engines, games and apps take turns without OOM crashes.

Tested on Linux

Use it

Linux x86_64

You need first

  • A GPU is required
  1. download berth binary and checksums

    curl -fL -o /tmp/berth-linux-amd64 https://github.com/T-Crypt/berth/releases/latest/download/berth-linux-amd64 && curl -fL -o /tmp/SHA256SUMS https://github.com/T-Crypt/berth/releases/latest/download/SHA256SUMS
  2. verify checksum and install berth

    (cd /tmp && sha256sum --check --ignore-missing SHA256SUMS) && install -Dm755 /tmp/berth-linux-amd64 ~/.local/bin/berth && berth version
Where the README and reality differ
  • README: install -Dm755 berth-linux-amd64 ~/.local/bin/berth
    What worked: curl -fL -o /tmp/berth-linux-amd64 https://github.com/T-Crypt/berth/releases/latest/download/berth-linux-amd64 plus SHA256SUMS, sha256sum --check, then install -Dm755
  • README: berth status shows the card
    What worked: start `berth daemon` (daemon socket /tmp/berth-1000.sock) before `berth status`; without it status exits 2 dial unix connect: no such file
  • README: berth gateway: one OpenAI-compatible port
    What worked: gateway /v1/models needs Authorization: Bearer $(cat ~/.config/berth/api.key) from `berth key`; unauthenticated curl fails; gateway log says startable=0 with no engines configured and returns empty list

Check that it works

berth --help

The output should contain “gateway one OpenAI-compatible port for every engine (preview)”.

First thing to try

nohup berth daemon > /tmp/berthd.log 2>&1 & sleep 3 && ls -l /tmp/berth-1000.sock
berth status

Run berth status to see the current state of GPU memory scheduling.

What it does

  • Let apps take turns sharing GPU memory
    Not shown working
  • See which app holds the GPU
    Not shown working
  • Start engines on demand through the gateway
    Not tested

How far pickfit got

Linux x86_64

  1. Installs
  2. Starts
  3. 0 of 3 features work
  4. Works offline

Windows: the project says it runs there; pickfit hasn’t tried it.

Good for

  • run an AI engine under a GPU lease
  • show who holds the GPU and who waits
  • give games the card while playing
  • download models from Hugging Face
  • view leases and telemetry in web UI

An alternative to llama-swap.

How pickfit tested it

Each run starts from a clean machine and follows the steps above. How pickfit verifies

Linux x86_64

Its main function works

  • Replayed from scratch
  • GPU: Tesla T4
  • Works offline
  • Checked on 1 Nov 2026
  • Version 0.4.0

How to read a tool card

Stitching
pickfit installed this tool itself. Everything under the stitches comes from that run. A tool pickfit has not run never gets stitches.
Verified on your OS
The install steps ran on the system you chose. “Ran on Linux only” means it has not been run on yours yet.
macOS, Linux and WindowsWhere it ran
Green OS marks: pickfit ran the install steps there. Grey ones: the README says it runs there, and pickfit has not tried.
How far the run got
Installs and starts, its main function works, its service answers, it produces model output. A window means a desktop app opened; a plug, an MCP client used it. The mark sits beside the run’s result, under “Why it fits and how we know” and in “How pickfit tested it”.
Visual check
pickfit looked at screenshots of the app. Green: it looks right. Amber with a number: things that look wrong; open it to see them. Grey and dashed: not checked yet.
Known vulnerabilities
Known high or critical vulnerabilities (OSV.dev) in what it installs. Green: none known. Red with a number: how many; open it for the list, and read it before you install. Grey and dashed: not checked, which does not mean safe.
About the run
The day it ran, and the version that ran.
Source of a reason
Every reason a pick fits rests on one profile field. Open the reason to see the field and the value it relies on.
Hidden gem
High quality, little known.
Quality and popularity
Two scores out of 100. Quality: how well it installed and worked when pickfit tried it, and how good its install docs and upkeep are. Popularity: from stars, downloads and Hacker News.
Features checked
Of the main features the README advertises, how many the run actually exercised (green), and how many failed (red). Grey ones were not checked, and the card says why.
Not verified
A well-known tool Fitz mentions from general knowledge. No run record, no steps, no score.