pickfit

BrowseLocal AI tools

kern

Model-agnostic GPU runtime that ships models as verified manifest packages.

Use it

Before you install: known vulnerabilities

pickfit found 2 known high or critical vulnerabilities in the packages a fresh install brings (OSV.dev, 3 Nov 2026). Weigh this before you install it.

  • crates:pulldown-cmark 0.13.4 Exponential event growth with `ENABLE_WIKILINKS` (RUSTSEC-2026-0414, high, fixed in 0.13.5)
See 1 more

Linux x86_64

You need first

  • Runtimes: python>=3.10
  • System packages: cuBLAS 13
  • A GPU is required
  1. pip install python runtime

    pip install kern-llm==0.3.1
  2. download install script

    curl -fsSL https://kern-baa.pages.dev/install.sh -o /tmp/kern-install.sh
  3. run install script for kern binaries

    sh /tmp/kern-install.sh
Where the README and reality differ
  • README: pip install kern-llm==0.3.1 provides kern usage
    What worked: uv tool install fails with 'No executables are provided by package kern-llm'; use pip install for Python runtime and install.sh for binaries
  • README: pip install kern-llm gives kern command
    What worked: pip install kern-llm==0.3.1 installs only Python import and no kern binary; run sh /tmp/kern-install.sh to get kern 0.3.1 and kern-serve

Check that it works

kern --version

The output should contain “kern 0.3.1 (48e0709, cuda 13.0)”.

First thing to try

git clone --depth 1 https://github.com/pegainfer-project/kern /tmp/kern-src && ls /tmp/kern-src/examples/ && ls /tmp/kern-src/*.toml
python3 -c "import json; d=json.load(open('/tmp/kern-src/examples/minimal.json')); d.pop('chat',None); json.dump(d, open('/tmp/minimal-nochek.json','w'), indent=2)" && kern verify /tmp/minimal-nochek.json

Run kern verify <manifest.json> to check a model manifest package.

What it does

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

  • Package models as verified programs
    Not tested
  • Serve models through an OpenAI-compatible server
    Not tested
  • Test candidate manifests against references
    Not tested
  • Benchmark programs on real workloads
    Not tested

How far pickfit got

Linux x86_64

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

Good for

  • generate text with a local model
  • serve a model via OpenAI-compatible server
  • test a manifest change for regressions

An alternative to vLLM and SGLang.

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 0.3.1

Differences the test machine hit

  • README: needs cuBLAS 13 on the loader path
    In the test: script warns no cuBLAS 13 on loader path; verify and cut work without it
  • README: examples manifests verify with released binary
    In the test: examples from main branch fail verify on 0.3.1 binary with unknown field chat and unknown variant temperature

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.