Instructions to use dheerubhadoria/Geog-Q4_K_M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use dheerubhadoria/Geog-Q4_K_M with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf dheerubhadoria/Geog-Q4_K_M:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf dheerubhadoria/Geog-Q4_K_M:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf dheerubhadoria/Geog-Q4_K_M:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf dheerubhadoria/Geog-Q4_K_M:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf dheerubhadoria/Geog-Q4_K_M:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf dheerubhadoria/Geog-Q4_K_M:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf dheerubhadoria/Geog-Q4_K_M:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf dheerubhadoria/Geog-Q4_K_M:Q4_K_M

Use Docker

docker model run hf.co/dheerubhadoria/Geog-Q4_K_M:Q4_K_M

LM Studio
Jan

vLLM

How to use dheerubhadoria/Geog-Q4_K_M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dheerubhadoria/Geog-Q4_K_M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dheerubhadoria/Geog-Q4_K_M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/dheerubhadoria/Geog-Q4_K_M:Q4_K_M

Ollama
How to use dheerubhadoria/Geog-Q4_K_M with Ollama:
```
ollama run hf.co/dheerubhadoria/Geog-Q4_K_M:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use dheerubhadoria/Geog-Q4_K_M with Docker Model Runner:
```
docker model run hf.co/dheerubhadoria/Geog-Q4_K_M:Q4_K_M
```

Lemonade

How to use dheerubhadoria/Geog-Q4_K_M with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull dheerubhadoria/Geog-Q4_K_M:Q4_K_M

Run and chat with the model

lemonade run user.Geog-Q4_K_M-Q4_K_M

List all available models

lemonade list

Atomic Chat

Geog-Q4_K_M

Geog-Q4_K_M Banner

Geog-Q4_K_M is a GGUF-format language model package intended for geography study and question-answering experiments. The project is designed around a SmolLM2-based workflow and geography-focused training data.

The model is packaged for llama.cpp-compatible runtimes and can be used locally with tools such as llama-cli, LM Studio, Ollama-compatible GGUF loaders, or other GGUF-supported inference apps.

Model Details

Model name: Geog-Q4_K_M
Format: GGUF
Quantization target: Q4_K_M
Base workflow: SmolLM2-style causal language model workflow
Domain: Geography
Language: English
Runtime target: llama.cpp-compatible inference

Model Statistics

Attribute	Value
Model Name	Geog-Q4_K_M
Base Model	SmolLM2
Parameters	135M
Domain	Geography
Language	English
Model Format	GGUF
Quantization	Q4_K_M
Training Samples	1,000
Dataset Format	JSON
Dataset Files	10
Average Samples per File	100
Dataset Schema	8 Structured Fields
Supported Topics	Earth, Climate, Maps, Latitude & Longitude, Physical Geography, Indian Geography, Natural Resources, and more
Inference Runtime	llama.cpp Compatible

Usage

llama-cli -m Geog-Q4_K_M.gguf -p "Explain latitude and longitude in simple words."

Example prompts:

What is the difference between weather and climate?
Explain the formation of monsoon winds.
Give a short note on plate tectonics.
Create 5 geography MCQs about rivers.

Training data

This project is set up around a custom geography dataset containing 1,000 structured records.

Each record includes structured fields such as:

category
section
topic
context
question
answer
difficulty
keywords

The dataset covers multiple geography areas, including topics such as Earth, latitude and longitude, physical geography, climate, natural resources, maps, Indian geography, and general geography study concepts.

Intended Use

This model package is intended for:

Geography study assistance
Short factual explanations
Topic summaries
Basic question-answering experiments
Local GGUF inference testing

Limitations

The model may produce incorrect or incomplete information.
The dataset is small compared with large-scale pretraining corpora.
Outputs should be verified before use in exams, teaching material, or decision-making.
This model is not a replacement for authoritative geography textbooks, atlases, government sources, or academic references.

Notes

This repository contains the upload-ready GGUF package and model card. The local project contains the scripts used to prepare data, train/export, and package the model.

Check and respect the license of any training data before publishing derived models.

Downloads last month: 16

GGUF

Model size

0.1B params

Architecture

llama

Hardware compatibility

4-bit

Model tree for dheerubhadoria/Geog-Q4_K_M

Base model

HuggingFaceTB/SmolLM2-135M

Quantized

(66)

this model