Instructions to use OpenAssistant/pythia-12b-sft-v8-7k-steps with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenAssistant/pythia-12b-sft-v8-7k-steps with Transformers:

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

pipe = pipeline("text-generation", model="OpenAssistant/pythia-12b-sft-v8-7k-steps")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("OpenAssistant/pythia-12b-sft-v8-7k-steps")
model = AutoModelForCausalLM.from_pretrained("OpenAssistant/pythia-12b-sft-v8-7k-steps", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use OpenAssistant/pythia-12b-sft-v8-7k-steps with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenAssistant/pythia-12b-sft-v8-7k-steps"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenAssistant/pythia-12b-sft-v8-7k-steps",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/OpenAssistant/pythia-12b-sft-v8-7k-steps

SGLang

How to use OpenAssistant/pythia-12b-sft-v8-7k-steps 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/pythia-12b-sft-v8-7k-steps" \
    --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/pythia-12b-sft-v8-7k-steps",
		"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/pythia-12b-sft-v8-7k-steps" \
        --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/pythia-12b-sft-v8-7k-steps",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use OpenAssistant/pythia-12b-sft-v8-7k-steps with Docker Model Runner:
```
docker model run hf.co/OpenAssistant/pythia-12b-sft-v8-7k-steps
```

Inference API

by gsaivinay - opened May 9, 2023

Discussion

gsaivinay

May 9, 2023

Hello,

can we replace the previous pythia model with this latest iteration for the inference API widgets?

julien-c

May 9, 2023

@sonatasv are you a member of the openassistant org? Thanks!

gsaivinay

May 9, 2023

•

edited May 17, 2023

@julien-c Hello,

No, I'm not. just looking for default inference API to play with the model.

julien-c

May 9, 2023

ok thanks! was just curious

gsaivinay

May 9, 2023

@julien-c - on a completely different note, I saw you mentioning that you are running huggingface chat (https://huggingface.co/spaces/huggingchat/chat-ui) backend server models in AWS g5 instances. Would it be okay for you to share the server launch configuration parameters? I'm also using g5 instances with quantization, but generation speed is not as good as huggingface chat.

Huggingface chat is lightning fast.

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