Instructions to use mgoin/Sheared-LLaMA-1.3B-ds with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mgoin/Sheared-LLaMA-1.3B-ds with Transformers:

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

pipe = pipeline("text-generation", model="mgoin/Sheared-LLaMA-1.3B-ds")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mgoin/Sheared-LLaMA-1.3B-ds")
model = AutoModelForCausalLM.from_pretrained("mgoin/Sheared-LLaMA-1.3B-ds", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use mgoin/Sheared-LLaMA-1.3B-ds with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mgoin/Sheared-LLaMA-1.3B-ds"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mgoin/Sheared-LLaMA-1.3B-ds",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mgoin/Sheared-LLaMA-1.3B-ds

SGLang

How to use mgoin/Sheared-LLaMA-1.3B-ds 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 "mgoin/Sheared-LLaMA-1.3B-ds" \
    --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": "mgoin/Sheared-LLaMA-1.3B-ds",
		"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 "mgoin/Sheared-LLaMA-1.3B-ds" \
        --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": "mgoin/Sheared-LLaMA-1.3B-ds",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mgoin/Sheared-LLaMA-1.3B-ds with Docker Model Runner:
```
docker model run hf.co/mgoin/Sheared-LLaMA-1.3B-ds
```

https://huggingface.co/princeton-nlp/Sheared-LLaMA-1.3B

from deepsparse import TextGeneration
model = TextGeneration(model="hf:mgoin/Sheared-LLaMA-1.3B-ds")
model("Say hello world in 10 languages:", max_new_tokens=100)
# TextGenerationOutput(created=datetime.datetime(2023, 10, 17, 22, 42, 52, 602506), prompts='Say hello world in 10 languages:', generations=[GeneratedText(text='\nThe world is a big place, and we want to make it a little bit smaller.\nWe’re building a platform that lets you say hello to the world in 10 languages.\nWe’re building a platform that lets you say hello to the world in 10 languages. We’re building a platform that lets you say hello to the world in 10 languages. We’re building a platform that lets you say hello to the world in 10 languages.', score=None, finished=True, finished_reason='length')])
model("Here is a recipe for banana bread:", max_new_tokens=100)
# TextGenerationOutput(created=datetime.datetime(2023, 10, 17, 22, 43, 57, 755015), prompts='Here is a recipe for banana bread:', generations=[GeneratedText(text='a simple, delicious, and healthy way to start your day.\nI’ve been making this banana bread for years. It’s a great way to start your day. It’s easy to make, and it’s delicious.\nI’ve made this banana bread for years. It’s a great way to start your day. It’s easy to make, and it’s delicious.\nI’ve made this banana bread for years', score=None, finished=True, finished_reason='length')])

Downloads last month: 9