Kev-4B by ggml-org

Kev-4B by ggml-org

Makes decisions through the /v1/systemone API of a llama server, from a GGUF copy of a 4.2B model. For developers who want a small local decision model.

Made by ggml-org on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: Kev-4B is a GGUF copy of a 4.2B text classification model, made by ggml-org. The card calls it a decision model, used through the /v1/systemone API. It was converted automatically from jaredpalmer/kev-4b, which builds on Qwen3.5-4B-Base.

Who it suits: It suits developers who want a small local model that makes a decision, rather than writing free text. The card gives few other details.

Title: Who runs it
Industry | Who uses it | How they use it | Example
Software | Developers | Call a local decision model through the systemone API | A routing step in an app
Research | Researchers testing llama tooling | Try a decision model served locally | Check how the endpoint behaves
IT | Teams keeping data in-house | Run a classifier on their own hardware | An internal triage helper

If you need a chat model that writes long answers, you can skip it.

Where to use it:

• The Llama app
• The llama command in a terminal on Windows, macOS or Linux
• Python on the same machines, for downloading the files by hand

Title: Jobs it does
Use case | Who | What they get back
Decisions through the API | Developers | A result from the /v1/systemone endpoint
Text classification | Researchers | A label for a piece of text

How to set it up:

• The Llama app or the llama command
• Python, only if you download the files by hand

1. Install the Llama app or the llama command.
2. Run the serve command below. It fetches the model by its Hugging Face name.
3. Optionally download the files by hand with the Hugging Face tools.

llama serve -hf ggml-org/Kev-4B-GGUF

pip install -U huggingface_hub
hf download ggml-org/Kev-4B-GGUF

How to use it: Start the server with the serve command, then call it through the /v1/systemone API, which the card names as the way to use this model. The card gives no request format, so check the llama tools.

llama serve -hf ggml-org/Kev-4B-GGUF

Example: You start the server with the serve command and send a request to the /v1/systemone endpoint. The card shows no sample request or reply, so the shape of the answer comes from the llama tools you use.

For newcomers and experts: If you are new, a decision model picks an answer from a set of options instead of writing a paragraph. The first small thing to try is the serve command above.

If you know the field, this is a 4.2B model that is about 2.1 GB at 4-bit by Hugging Face's estimate, with Qwen3.5-4B-Base as an upstream source. The card gives no benchmark, label set or request schema, so treat it as a converted checkpoint and test it yourself.

Run llama serve with the model name
Files come from Hugging Face
The server starts on your machine
Call the /v1/systemone endpoint

Good to know:

• Conversion: an automatic tool from ggml-org made this copy from the source model.
• Sources: jaredpalmer/kev-4b and Qwen3.5-4B-Base.
• Detail: the card has no scores or usage notes beyond the endpoint name.
• License: apache-2.0.
• Cost to run: no API key, account or paid service is named.
• Data it reaches: you download the model from Hugging Face, and the card names no other access.

On Hugging Face, 13 November 2026

• Task: text classification
• Parameters: 4.2B
• Weights: about 2.1 GB at 4-bit (estimate)
• Formats: GGUF
• Downloads in the last 30 days: 39,531
• Likes: 34
• License: apache-2.0
• Based on: jaredpalmer/kev-4b
• Tags: gguf, quantized, decision-model, text-classification, conversational

Links

• Model: https://huggingface.co/ggml-org/Kev-4B-GGUF
• Files: https://huggingface.co/ggml-org/Kev-4B-GGUF/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download ggml-org/Kev-4B-GGUF

Then follow the usage in its model card. Nothing in this kit runs by itself.