Instructions to use joongi007/JEJUMA-002-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use joongi007/JEJUMA-002-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="joongi007/JEJUMA-002-GGUF",
	filename="JEJUMA-002-Q2_K.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use joongi007/JEJUMA-002-GGUF 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 joongi007/JEJUMA-002-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf joongi007/JEJUMA-002-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf joongi007/JEJUMA-002-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf joongi007/JEJUMA-002-GGUF: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 joongi007/JEJUMA-002-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf joongi007/JEJUMA-002-GGUF: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 joongi007/JEJUMA-002-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf joongi007/JEJUMA-002-GGUF:Q4_K_M

Use Docker

docker model run hf.co/joongi007/JEJUMA-002-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use joongi007/JEJUMA-002-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "joongi007/JEJUMA-002-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "joongi007/JEJUMA-002-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/joongi007/JEJUMA-002-GGUF:Q4_K_M

Ollama
How to use joongi007/JEJUMA-002-GGUF with Ollama:
```
ollama run hf.co/joongi007/JEJUMA-002-GGUF:Q4_K_M
```

Unsloth Studio

How to use joongi007/JEJUMA-002-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for joongi007/JEJUMA-002-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for joongi007/JEJUMA-002-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for joongi007/JEJUMA-002-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use joongi007/JEJUMA-002-GGUF with Docker Model Runner:
```
docker model run hf.co/joongi007/JEJUMA-002-GGUF:Q4_K_M
```

Lemonade

How to use joongi007/JEJUMA-002-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull joongi007/JEJUMA-002-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.JEJUMA-002-GGUF-Q4_K_M

List all available models

lemonade list

JEJUMA-002-GGUF / README.md

joongi007

Update README.md

a71705b verified almost 2 years ago

preview code

Raw

History Blame Contribute Delete

1.88 kB

metadata

license: mit
base_model: JEJUMA/JEJUMA-002
tags:
  - gguf
  - Dialect
  - Language
model-index:
  - name: joongi007/JEJUMA-002-GGUF
    results: []
language:
  - ko
pipeline_tag: text-generation

Original model is JEJUMA/JEJUMA-002 - bbd7ec2
quantized using llama.cpp - b3542
JEJUMA Official Quantization is JEJUMA/JEJUMA-002-GGUF
After trying out this model, I noticed a few things:
1. It's more like a translation model. You can't chat with it, it only does translations.
2. It can only handle one dialect (or standard Korean) at a time.
3. Don't expect a conversation. It's strictly for translation purposes! Look at the example below!
  system prompt
```
Answer your questions using the Jeju dialect.
```
  user question
```
hello! How are you doing now?
```
  assistant answer
```
헐쯤 험과게
# 할수 많습니까
```

Prompt(LM Studio)

<|start_header_id|>system<|end_header_id|>

{System}
<|eot_id|><|start_header_id|>user<|end_header_id|>

{User}
<|eot_id|><|start_header_id|>assistant<|end_header_id|>

{Assistant}

Example of User Prompts

Detect the following sentence or word is standard, jeju, chungcheong, gangwon, gyeongsang, or jeonla's dialect: 
```
{Enter the Jeju island dialect or standard Korean here}
```

Detect the following sentence or word is which dialect and convert the following sentence or word to standard Korean: 
```
{Enter Jeju island dialect or standard Korean here}
```