pickfit

BrowseLocal AI tools

PocketLLM

Single-device inference engine that runs Qwen3 text, vision and speech models via a C core.

Tested on Linux

Use it

Linux x86_64

You need first

  • Runtimes: python>=3.10
  1. clone PocketLLM source from GitHub

    git clone --depth 1 https://github.com/lvyufeng/PocketLLM.git /tmp/PocketLLM
  2. pip install PocketLLM from cloned source

    pip install --user /tmp/PocketLLM 2>&1 && python3 -m pocketllm --help
Where the README and reality differ
  • README: pip install pocketllm then `pocketllm devices`
    What worked: clone the GitHub tree and run `pip install --user /tmp/PocketLLM`, then invoke as `python3 -m pocketllm devices` (no `pocketllm` executable is shipped; pyproject.toml defines no [project.scripts])
  • README: pip install pocketllm installs the documented tool
    What worked: PyPI pocketllm 0.1.1 fails to build in isolation: setup.py defines `class BuildExtensions(BuildExtension)` with BuildExtension=None when torch is absent, raising `TypeError: NoneType takes no arguments`; the current GitHub tree (0.2.0.dev0, numpy-only, no ext_modules) installs cleanly

Check that it works

python3 -m pocketllm --help 2>&1

The output should contain “usage: pocketllm”.

First thing to try

python3 -m pocketllm devices 2>&1

Run python3 -m pocketllm --help to see the available commands.

What it does

These come from the README; pickfit didn’t check them one by one.

  • Run large language models on a single accelerator
    Not tested
  • Serve models over an OpenAI-compatible HTTP API
    Not tested
  • Transcribe speech audio to text
    Not tested
  • Process images with a vision-language model
    Not tested

How far pickfit got

Linux x86_64

  1. Installs
  2. Starts
  3. Main function works
  4. Works offline

Good for

  • generate text from a Qwen3 GGUF checkpoint via CLI
  • serve model behind OpenAI-compatible HTTP
  • transcribe speech to transcript with Qwen3-ASR

An alternative to llama.cpp.

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
  • Works offline
  • Checked on 30 Oct 2026
  • Version 0.2.0.dev0

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.