Instructions to use svjack/prompt-extend-chinese-gpt with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use svjack/prompt-extend-chinese-gpt with Transformers:

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

pipe = pipeline("text-generation", model="svjack/prompt-extend-chinese-gpt")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("svjack/prompt-extend-chinese-gpt")
model = AutoModelForMultimodalLM.from_pretrained("svjack/prompt-extend-chinese-gpt")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use svjack/prompt-extend-chinese-gpt with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/svjack/prompt-extend-chinese-gpt

SGLang

How to use svjack/prompt-extend-chinese-gpt 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 "svjack/prompt-extend-chinese-gpt" \
    --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": "svjack/prompt-extend-chinese-gpt",
		"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 "svjack/prompt-extend-chinese-gpt" \
        --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": "svjack/prompt-extend-chinese-gpt",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use svjack/prompt-extend-chinese-gpt with Docker Model Runner:
```
docker model run hf.co/svjack/prompt-extend-chinese-gpt
```

from transformers import BertTokenizer, GPT2LMHeadModel
device = "cpu"

model_path = "svjack/prompt-extend-chinese-gpt"
tokenizer = BertTokenizer.from_pretrained(model_path)
model = GPT2LMHeadModel.from_pretrained(model_path)

prompt = "一只凶猛的老虎，咬死了一只豺狼。"

encode = tokenizer(prompt, return_tensors='pt').to(device)
answer = model.generate(encode.input_ids,
                       max_length = 128,
    num_beams=2,
    top_p = 0.95,
    top_k = 50,
    repetition_penalty = 2.5,
    length_penalty=1.0,
    early_stopping=True,
                       )[0]
decoded = tokenizer.decode(answer, skip_special_tokens=True)
decoded

'一 只 凶 猛 的 老 虎 ， 咬 死 了 一 只 豺 狼 。 高 度 详 细 ， 数 字 绘 画 ， 艺 术 站 ， 概 念 艺 术 ， 锐 利 的 焦 点 ， 插 图 ， 电 影 照 明 ， 艺 术 由 artgerm 和 greg rutkowski 和 alphonse mucha 8 k 彩 色 辛 烷 渲 染 。'

Downloads last month: 7

svjack
/

prompt-extend-chinese-gpt

Spaces using svjack/prompt-extend-chinese-gpt 2