Instructions to use Yukang/LongAlpaca-70B-16k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Yukang/LongAlpaca-70B-16k with Transformers:

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

pipe = pipeline("text-generation", model="Yukang/LongAlpaca-70B-16k")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Yukang/LongAlpaca-70B-16k")
model = AutoModelForCausalLM.from_pretrained("Yukang/LongAlpaca-70B-16k", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Yukang/LongAlpaca-70B-16k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Yukang/LongAlpaca-70B-16k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Yukang/LongAlpaca-70B-16k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Yukang/LongAlpaca-70B-16k

SGLang

How to use Yukang/LongAlpaca-70B-16k 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 "Yukang/LongAlpaca-70B-16k" \
    --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": "Yukang/LongAlpaca-70B-16k",
		"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 "Yukang/LongAlpaca-70B-16k" \
        --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": "Yukang/LongAlpaca-70B-16k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Yukang/LongAlpaca-70B-16k with Docker Model Runner:
```
docker model run hf.co/Yukang/LongAlpaca-70B-16k
```

Thank you

by MB7977 - opened Nov 17, 2023

Discussion

MB7977

Nov 17, 2023

Thanks so much for training a 16K version. The LongLoRA approach is the best I've worked with in open models, really appreciate your work. I presume this uses the Llama 2 prompt format like the updated Alpaca models?

Yukang

Owner Nov 18, 2023

Hi,

Many thanks for your comment. For this model, its prompt format also follows the Llama2 prompt format. And we do not include system prompt for supervised fine-tuning.
You can simply use the line below for it.

f"[INST]{instruction}[/INST]"

Regards,
Yukang Chen

MB7977

Nov 18, 2023

Thank you.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment