Instructions to use OpenAssistant/oasst-sft-1-pythia-12b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenAssistant/oasst-sft-1-pythia-12b with Transformers:

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

pipe = pipeline("text-generation", model="OpenAssistant/oasst-sft-1-pythia-12b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("OpenAssistant/oasst-sft-1-pythia-12b")
model = AutoModelForCausalLM.from_pretrained("OpenAssistant/oasst-sft-1-pythia-12b")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use OpenAssistant/oasst-sft-1-pythia-12b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenAssistant/oasst-sft-1-pythia-12b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenAssistant/oasst-sft-1-pythia-12b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/OpenAssistant/oasst-sft-1-pythia-12b

SGLang

How to use OpenAssistant/oasst-sft-1-pythia-12b 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/oasst-sft-1-pythia-12b" \
    --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/oasst-sft-1-pythia-12b",
		"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/oasst-sft-1-pythia-12b" \
        --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/oasst-sft-1-pythia-12b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use OpenAssistant/oasst-sft-1-pythia-12b with Docker Model Runner:
```
docker model run hf.co/OpenAssistant/oasst-sft-1-pythia-12b
```

Where is the raw data?

by Tom-nerd - opened Apr 4, 2023

Discussion

Tom-nerd

Apr 4, 2023

I appreciate putting out RLHF model alternatives. Though the raw data would be great, maybe including the ratings of those strings.

OllieStanley

OpenAssistant org Apr 4, 2023

It was announced a while ago that the data will be released alongside the first version of the model on April 15th

BenBot

Apr 7, 2023

@OllieStanley Do you have a link to that announcement? I must've missed it.

OllieStanley

OpenAssistant org Apr 7, 2023

@OllieStanley Do you have a link to that announcement? I must've missed it.

It's in the Open Assistant Discord announcements channel

Tom-nerd

Apr 7, 2023

@OllieStanley Do you have a link to that announcement? I must've missed it.

It's in the Open Assistant Discord announcements channel

I dont know where he got that. But I do know that 15 hours ago Yannic kilcher announced it in this video: https://www.youtube.com/watch?v=Hi6cbeBY2oQ

OllieStanley

OpenAssistant org Apr 7, 2023

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