Instructions to use OpenAssistant/llama2-13b-orca-8k-3319 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenAssistant/llama2-13b-orca-8k-3319 with Transformers:

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

pipe = pipeline("text-generation", model="OpenAssistant/llama2-13b-orca-8k-3319")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("OpenAssistant/llama2-13b-orca-8k-3319")
model = AutoModelForCausalLM.from_pretrained("OpenAssistant/llama2-13b-orca-8k-3319")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use OpenAssistant/llama2-13b-orca-8k-3319 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenAssistant/llama2-13b-orca-8k-3319"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenAssistant/llama2-13b-orca-8k-3319",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/OpenAssistant/llama2-13b-orca-8k-3319

SGLang

How to use OpenAssistant/llama2-13b-orca-8k-3319 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 "OpenAssistant/llama2-13b-orca-8k-3319" \
    --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": "OpenAssistant/llama2-13b-orca-8k-3319",
		"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 "OpenAssistant/llama2-13b-orca-8k-3319" \
        --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": "OpenAssistant/llama2-13b-orca-8k-3319",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use OpenAssistant/llama2-13b-orca-8k-3319 with Docker Model Runner:
```
docker model run hf.co/OpenAssistant/llama2-13b-orca-8k-3319
```

Awesome, thanks

by gsaivinay - opened Jul 25, 2023

Discussion

gsaivinay

Jul 25, 2023

Thanks for this model with 8k context 🤯

base model: meta-llama/Llama-2-7b

is this a typo in Readme?

Also I'd like to ask few things:

Is this model trained with any coding data? given that it has 8k context, it'll be excellent use case for coding assistant.
Is there any prompt format for this model to answer questions based on context? for example using it for document chat using vector db.
did you happen to benchmark this model? while I can wait for open llm leaderboard, it may take few days for this model to be evaluated in the leaderboard.

andreaskoepf

OpenAssistant org Jul 25, 2023

base model: "meta-llama/Llama-2-7b" - is this a typo in Readme?

Yes .. good catch, it is of course based on Llama-2-13b, should be fixed now!

Is this model trained with any coding data? given that it has 8k context, it'll be excellent use case for coding assistant.

From benchmarks we know that it is currently not performing well in coding tasks. We are further fine-tuning the model with more code related instruction data.

Is there any prompt format for this model to answer questions based on context? for example using it for document chat using vector db.

No, the context would probably best be placed in the <|prompter|> message.

did you happen to benchmark this model? while I can wait for open llm leaderboard, it may take few days for this model to be evaluated in the leaderboard.

Some benchmark results can be found here: https://tju01.github.io/ilm-eval/#?branch=oa-orca

gsaivinay changed discussion status to closed Jul 26, 2023

gsaivinay

Jul 26, 2023

evaluation available now

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