Instructions to use VMware/open-llama-7b-v2-open-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use VMware/open-llama-7b-v2-open-instruct with Transformers:

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

pipe = pipeline("text-generation", model="VMware/open-llama-7b-v2-open-instruct")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("VMware/open-llama-7b-v2-open-instruct")
model = AutoModelForCausalLM.from_pretrained("VMware/open-llama-7b-v2-open-instruct")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use VMware/open-llama-7b-v2-open-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "VMware/open-llama-7b-v2-open-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "VMware/open-llama-7b-v2-open-instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/VMware/open-llama-7b-v2-open-instruct

SGLang

How to use VMware/open-llama-7b-v2-open-instruct 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 "VMware/open-llama-7b-v2-open-instruct" \
    --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": "VMware/open-llama-7b-v2-open-instruct",
		"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 "VMware/open-llama-7b-v2-open-instruct" \
        --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": "VMware/open-llama-7b-v2-open-instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use VMware/open-llama-7b-v2-open-instruct with Docker Model Runner:
```
docker model run hf.co/VMware/open-llama-7b-v2-open-instruct
```

Impressive

by captainst - opened Jul 12, 2023

Discussion

captainst

Jul 12, 2023

Really impressive. The model is able to answer questions based on the software log input of about 2K tokens.
My first test shows that the performance is comparable to bigger models like 13b ones, and sometimes even better.

rbattle

VMware AI Labs org Jul 12, 2023

Awesome, thank you for letting us know!

Hao-Hu

VMware AI Labs org Jul 19, 2023

Model performance comparison with before version on the MODEL CARD: https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md

rbattle

VMware AI Labs org Jul 19, 2023

This model is built on OpenLLaMA, not Llama 2. It was fine-tuned on version 2 of our open-instruct training set.

haohoo

Jul 20, 2023

This model is built on OpenLLaMA, not Llama 2. It was fine-tuned on version 2 of our open-instruct training set.

Oh sorry, I mixed them. Just because LLaMA2 is too hot yesterday.

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