Instructions to use MrBananaHuman/kogpt_6b_fp16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MrBananaHuman/kogpt_6b_fp16 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="MrBananaHuman/kogpt_6b_fp16")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MrBananaHuman/kogpt_6b_fp16")
model = AutoModelForCausalLM.from_pretrained("MrBananaHuman/kogpt_6b_fp16")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use MrBananaHuman/kogpt_6b_fp16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MrBananaHuman/kogpt_6b_fp16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MrBananaHuman/kogpt_6b_fp16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/MrBananaHuman/kogpt_6b_fp16

SGLang

How to use MrBananaHuman/kogpt_6b_fp16 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "MrBananaHuman/kogpt_6b_fp16" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MrBananaHuman/kogpt_6b_fp16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "MrBananaHuman/kogpt_6b_fp16" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MrBananaHuman/kogpt_6b_fp16",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use MrBananaHuman/kogpt_6b_fp16 with Docker Model Runner:
```
docker model run hf.co/MrBananaHuman/kogpt_6b_fp16
```

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

kakao brain에서 공개한 kogpt 6b model('kakaobrain/kogpt')을 fp16으로 저장한 모델입니다.

카카오브레인 모델을 fp16으로 로드하는 방법

import torch
from transformers import GPTJForCausalLM

model = GPTJForCausalLM.from_pretrained('kakaobrain/kogpt', cache_dir='./my_dir', revision='KoGPT6B-ryan1.5b', torch_dtype=torch.float16)

fp16 모델 로드 후 문장 생성

import torch
from transformers import GPTJForCausalLM, AutoTokenizer

model = GPTJForCausalLM.from_pretrained('MrBananaHuman/kogpt_6b_fp16', low_cpu_mem_usage=True))
model.to('cuda')
tokenizer = AutoTokenizer.from_pretrained('MrBananaHuman/kogpt_6b_fp16')

input_text = '이순신은'
input_ids = tokenizer(input_text, return_tensors='pt').input_ids.to('cuda')

output = model.generate(input_ids, max_length=64)
print(tokenizer.decode(output[0]))

>>> 이순신은 우리에게 무엇인가? 1. 머리말 이글은 임진왜란 당시 이순인이 보여준

참고 링크

https://github.com/kakaobrain/kogpt/issues/6?fbclid=IwAR1KpWhuHnevQvEWV18o16k2z9TLgrXkbWTkKqzL-NDXHfDnWcIq7I4SJXM

Downloads last month: 10

MrBananaHuman
/

kogpt_6b_fp16

카카오브레인 모델을 fp16으로 로드하는 방법

fp16 모델 로드 후 문장 생성

참고 링크

Space using MrBananaHuman/kogpt_6b_fp16 1