Instructions to use QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF", dtype="auto")

llama-cpp-python

How to use QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF",
	filename="llama-3.2-Korean-Bllossom-3B.Q2_K.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf QuantFactory/llama-3.2-Korean-Bllossom-3B-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 QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf QuantFactory/llama-3.2-Korean-Bllossom-3B-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 QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF with Ollama:
```
ollama run hf.co/QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M
```

Unsloth Studio

How to use QuantFactory/llama-3.2-Korean-Bllossom-3B-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 QuantFactory/llama-3.2-Korean-Bllossom-3B-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 QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF to start chatting

How to use QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF with Docker Model Runner:
```
docker model run hf.co/QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M
```

Lemonade

How to use QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.llama-3.2-Korean-Bllossom-3B-GGUF-Q4_K_M

List all available models

lemonade list

QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF

This is quantized version of Bllossom/llama-3.2-Korean-Bllossom-3B created using llama.cpp

Original Model Card

Update!

[2024.10.08] Bllossom-3B 모델이 최초 업데이트 되었습니다.

Bllossom | Demo | Homepage | Github |

저희 Bllossom 팀에서 Bllossom-3B 모델을 공개합니다.
llama3.2-3B가 나왔는데 한국어가 포함 안되었다구?? 이번 Bllossom-3B는 한국어가 지원되지 않는 기본 모델을 한국어-영어로 강화모델입니다.
 - 100% full-tuning으로 150GB의 정제된 한국어로 추가 사전학습 되었습니다. (GPU많이 태웠습니다)
 - 굉장히 정제된 Instruction Tuning을 진행했습니다.
 - 영어 성능을 전혀 손상시키지 않은 완전한 Bilingual 모델입니다.
 - LogicKor 기준 5B이하 최고점수를 기록했고 6점 초반대 점수를 보입니다.
 - Instruction tuning만 진행했습니다. DPO 등 성능 올릴 방법으로 튜닝해보세요.
 - MT-Bench, LogicKor 등 벤치마크 점수를 잘받기 위해 정답데이터를 활용하거나 혹은 벤치마크를 타겟팅 해서 학습하지 않았습니다. (해당 벤치마크 타게팅해서 학습하면 8점도 나옵니다...)

언제나 그랬듯 해당 모델은 상업적 이용이 가능합니다.

1. Bllossom은 AAAI2024, NAACL2024, LREC-COLING2024 (구두) 발표되었습니다.
2. 좋은 언어모델 계속 업데이트 하겠습니다!! 한국어 강화를위해 공동 연구하실분(특히논문) 언제든 환영합니다!!

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = 'Bllossom/llama-3.2-Korean-Bllossom-3B'

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)
instruction = "철수가 20개의 연필을 가지고 있었는데 영희가 절반을 가져가고 민수가 남은 5개를 가져갔으면 철수에게 남은 연필의 갯수는 몇개인가요?"

messages = [
    {"role": "user", "content": f"{instruction}"}
    ]

input_ids = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    return_tensors="pt"
).to(model.device)

terminators = [
    tokenizer.convert_tokens_to_ids("<|end_of_text|>"),
    tokenizer.convert_tokens_to_ids("<|eot_id|>")
]

outputs = model.generate(
    input_ids,
    max_new_tokens=1024,
    eos_token_id=terminators,
    do_sample=True,
    temperature=0.6,
    top_p=0.9
)

print(tokenizer.decode(outputs[0][input_ids.shape[-1]:], skip_special_tokens=True))

철수가 20개의 연필을 가지고 있었고 영희가 절반을 가져가면, 영희가 가져간 연필의 갯수는 20 / 2 = 10개입니다.

이제 철수가 남은 연필의 갯수를 계산해보겠습니다. 영희가 10개를 가져간 후 철수가 남은 연필의 갯수는 20 - 10 = 10개입니다.

민수가 남은 5개를 가져갔으므로, 철수가 남은 연필의 갯수는 10 - 5 = 5개입니다. 

따라서 철수가 남은 연필의 갯수는 5개입니다.

Supported by

AICA

Citation

Language Model

@misc{bllossom,
  author = {ChangSu Choi, Yongbin Jeong, Seoyoon Park, InHo Won, HyeonSeok Lim, SangMin Kim, Yejee Kang, Chanhyuk Yoon, Jaewan Park, Yiseul Lee, HyeJin Lee, Younggyun Hahm, Hansaem Kim, KyungTae Lim},
  title = {Optimizing Language Augmentation for Multilingual Large Language Models: A Case Study on Korean},
  year = {2024},
  journal = {LREC-COLING 2024},
  paperLink = {\url{https://arxiv.org/pdf/2403.10882}},
 },
}

Vision-Language Model

@misc{bllossom-V,
  author = {Dongjae Shin, Hyunseok Lim, Inho Won, Changsu Choi, Minjun Kim, Seungwoo Song, Hangyeol Yoo, Sangmin Kim, Kyungtae Lim},
  title = {X-LLaVA: Optimizing Bilingual Large Vision-Language Alignment},
  year = {2024},
  publisher = {GitHub},
  journal = {NAACL 2024 findings},
  paperLink = {\url{https://arxiv.org/pdf/2403.11399}},
 },
}

Contact

임경태(KyungTae Lim), Professor at Seoultech. ktlim@seoultech.ac.kr
함영균(Younggyun Hahm), CEO of Teddysum. hahmyg@teddysum.ai
김한샘(Hansaem Kim), Professor at Yonsei. khss@yonsei.ac.kr

Contributor

유한결(Hangyeol Yoo), 21102372@seoultech.ac.kr
신동재(Dongjae Shin), dylan1998@seoultech.ac.kr
임현석(Hyeonseok Lim), gustjrantk@seoultech.ac.kr
원인호(Inho Won), wih1226@seoultech.ac.kr
김민준(Minjun Kim), mjkmain@seoultech.ac.kr
송승우(Seungwoo Song), sswoo@seoultech.ac.kr
육정훈(Jeonghun Yuk), usually670@gmail.com
최창수(Chansu Choi), choics2623@seoultech.ac.kr
송서현(Seohyun Song), alexalex225225@gmail.com

Downloads last month: 99

GGUF

Model size

3B params

Architecture

llama

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Papers for QuantFactory/llama-3.2-Korean-Bllossom-3B-GGUF

X-LLaVA: Optimizing Bilingual Large Vision-Language Alignment

Paper • 2403.11399 • Published Mar 18, 2024 • 6

Optimizing Language Augmentation for Multilingual Large Language Models: A Case Study on Korean

Paper • 2403.10882 • Published Mar 16, 2024 • 6