Instructions to use TheBloke/StableBeluga2-70B-GGML with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheBloke/StableBeluga2-70B-GGML with Transformers:

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

pipe = pipeline("text-generation", model="TheBloke/StableBeluga2-70B-GGML")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("TheBloke/StableBeluga2-70B-GGML", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use TheBloke/StableBeluga2-70B-GGML with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "TheBloke/StableBeluga2-70B-GGML"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "TheBloke/StableBeluga2-70B-GGML",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/TheBloke/StableBeluga2-70B-GGML

SGLang

How to use TheBloke/StableBeluga2-70B-GGML 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 "TheBloke/StableBeluga2-70B-GGML" \
    --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": "TheBloke/StableBeluga2-70B-GGML",
		"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 "TheBloke/StableBeluga2-70B-GGML" \
        --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": "TheBloke/StableBeluga2-70B-GGML",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use TheBloke/StableBeluga2-70B-GGML with Docker Model Runner:
```
docker model run hf.co/TheBloke/StableBeluga2-70B-GGML
```

Performance of quantified models

by danielus - opened Jul 29, 2023

Discussion

danielus

Jul 29, 2023

Is there any way to work out how much 'performance' the quantised versions lose compared to the original? So that you can get an idea of which quantisation level to choose and maximise the ratio of resources used/generation performance
In the github repository of llama.cpp I only found an old post of the different accuracies for the quantisation levels, but I assume it has become obsolete given the speed at which this world is advancing!

jeffwadsworth

Aug 23, 2023

Trust me, the 8 bit is worth it. At least in regards to reasoning.

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