Instructions to use KnutJaegersberg/LongWriter-8b-exl-8.0bpw with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KnutJaegersberg/LongWriter-8b-exl-8.0bpw with Transformers:

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

pipe = pipeline("text-generation", model="KnutJaegersberg/LongWriter-8b-exl-8.0bpw")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("KnutJaegersberg/LongWriter-8b-exl-8.0bpw")
model = AutoModelForCausalLM.from_pretrained("KnutJaegersberg/LongWriter-8b-exl-8.0bpw")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use KnutJaegersberg/LongWriter-8b-exl-8.0bpw with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KnutJaegersberg/LongWriter-8b-exl-8.0bpw"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KnutJaegersberg/LongWriter-8b-exl-8.0bpw",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/KnutJaegersberg/LongWriter-8b-exl-8.0bpw

SGLang

How to use KnutJaegersberg/LongWriter-8b-exl-8.0bpw 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 "KnutJaegersberg/LongWriter-8b-exl-8.0bpw" \
    --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": "KnutJaegersberg/LongWriter-8b-exl-8.0bpw",
		"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 "KnutJaegersberg/LongWriter-8b-exl-8.0bpw" \
        --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": "KnutJaegersberg/LongWriter-8b-exl-8.0bpw",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use KnutJaegersberg/LongWriter-8b-exl-8.0bpw with Docker Model Runner:
```
docker model run hf.co/KnutJaegersberg/LongWriter-8b-exl-8.0bpw
```

LongWriter-llama3.1-8b

🤗 [LongWriter Dataset] • 💻 [Github Repo] • 📃 [LongWriter Paper]

LongWriter-llama3.1-8b is trained based on Meta-Llama-3.1-8B, and is capable of generating 10,000+ words at once.

A simple demo for deployment of the model:

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
tokenizer = AutoTokenizer.from_pretrained("THUDM/LongWriter-llama3.1-8b", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("THUDM/LongWriter-llama3.1-8b", torch_dtype=torch.bfloat16, trust_remote_code=True, device_map="auto")
model = model.eval()
query = "Write a 10000-word China travel guide"
prompt = f"[INST]{query}[/INST]"
input = tokenizer(prompt, truncation=False, return_tensors="pt").to(device)
context_length = input.input_ids.shape[-1]
output = model.generate(
    **input,
    max_new_tokens=32768,
    num_beams=1,
    do_sample=True,
    temperature=0.5,
)[0]
response = tokenizer.decode(output[context_length:], skip_special_tokens=True)
print(response)

License: Llama-3.1 License

Citation

If you find our work useful, please consider citing LongWriter:

Downloads last month: 2

Dataset used to train KnutJaegersberg/LongWriter-8b-exl-8.0bpw

Paper for KnutJaegersberg/LongWriter-8b-exl-8.0bpw

LongWriter: Unleashing 10,000+ Word Generation from Long Context LLMs

Paper • 2408.07055 • Published Aug 13, 2024 • 69