Laya-GGUF

Laya-GGUF

Small 421M-parameter decision model in GGUF form that you serve locally and call through a systemone API. For developers who use llama.cpp tooling and want a compact classifier.

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: Laya-GGUF is a GGUF conversion of Laya, a small decision model with 421 million parameters, filed under text classification. A conversion tool made the files automatically from the original model. The card is short, so it says little about what the model decides.

Who it suits: Developers who already use llama.cpp tools and want a compact model behind a local server.

Industry | Who uses it | How they use it | Example
Software | Application developers | Run a small decision model beside their app | Start the server and call its systemone API
Open-source tooling | llama.cpp users | Try a converted model with one command | Serve the model straight from Hugging Face
Education | Students | Look at how a small converted model is packaged | Compare its size with larger models

Skip it if you need documented labels or outputs, because the card does not describe them.

Where to use it:

• The Llama app
• A terminal, using the llama command
• Any application that can call the model's systemone API on a local server

Use case | Who | What they get back
Serve the model locally | Developers | A local server that hosts the model
Call the decision API | Application authors | Responses from the systemone API the card names

How to set it up: You need either the Llama app or its terminal command, plus a connection that can pull the model from Hugging Face. No key or account is named.

• The Llama app, or the llama command in a terminal
• Disk space for a small file, about 211 MB at 4-bit by the record's estimate

1. Install the Llama app or its terminal tool.
2. Run the serve command, which pulls the model from Hugging Face.

llama serve -hf ggml-org/Laya-GGUF

If you prefer to fetch the files yourself, use the hub tool.

pip install -U huggingface_hub
hf download ggml-org/Laya-GGUF

How to use it: Start the server with the serve command, then send requests to its systemone API. The card says this is a decision model meant for that API, and points to a llama.cpp change for details.

llama serve -hf ggml-org/Laya-GGUF

Tool | What it does
llama serve | Starts a local server and loads the model
-hf | Names the Hugging Face repository to pull the model from
systemone API | The interface the card says this model is used through

Example: You start the server with the serve command and send a request to its systemone API. The card shows no sample request or reply, so read the llama.cpp notes it points to for the exact format.

For newcomers and experts: If you are new to this, run the serve command and check that the server starts. That alone confirms the model loads on your machine.

If you know the field, you have a 421M-parameter GGUF file of a decision model, about 211 MB at 4-bit by estimate. It is not a chat model. Its request and response schema is documented outside the card.

Run the serve command
The model downloads and loads
Send a request to the systemone API
Read the model's decision

Good to know: The files were converted automatically, so training details and limits are not in the card.

• License: Apache 2.0
• Cost to run: none named
• Data it reaches: files pulled from Hugging Face, plus the requests you send to your own local server
• Not for: general chat, since the card calls it a decision model

On Hugging Face, 13 November 2026

• Task: text classification
• Parameters: 421M
• Weights: about 211 MB at 4-bit (estimate)
• Formats: GGUF
• Downloads in the last 30 days: 16,039
• Likes: 41
• License: apache-2.0
• Based on: convaiinnovations/laya
• Quantized versions: 1
• Tags: gguf, quantized, decision-model, text-classification, feature-extraction

Links

• Model: https://huggingface.co/ggml-org/Laya-GGUF
• Files: https://huggingface.co/ggml-org/Laya-GGUF/tree/main

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download ggml-org/Laya-GGUF

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