Not-For-All-Audiences

Instructions to use Dracones/perky-103b-v0.1_exl2_3.35bpw with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Dracones/perky-103b-v0.1_exl2_3.35bpw with Transformers:

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

pipe = pipeline("text-generation", model="Dracones/perky-103b-v0.1_exl2_3.35bpw")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("Dracones/perky-103b-v0.1_exl2_3.35bpw")
model = AutoModelForMultimodalLM.from_pretrained("Dracones/perky-103b-v0.1_exl2_3.35bpw")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Dracones/perky-103b-v0.1_exl2_3.35bpw with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Dracones/perky-103b-v0.1_exl2_3.35bpw"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Dracones/perky-103b-v0.1_exl2_3.35bpw",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Dracones/perky-103b-v0.1_exl2_3.35bpw

SGLang

How to use Dracones/perky-103b-v0.1_exl2_3.35bpw 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 "Dracones/perky-103b-v0.1_exl2_3.35bpw" \
    --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": "Dracones/perky-103b-v0.1_exl2_3.35bpw",
		"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 "Dracones/perky-103b-v0.1_exl2_3.35bpw" \
        --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": "Dracones/perky-103b-v0.1_exl2_3.35bpw",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Dracones/perky-103b-v0.1_exl2_3.35bpw with Docker Model Runner:
```
docker model run hf.co/Dracones/perky-103b-v0.1_exl2_3.35bpw
```

perky-103b-v0.1_exl2_3.35bpw / Perky.card.png

Dracones

Upload folder using huggingface_hub

f26dc1b verified over 2 years ago

download

history blame contribute delete

1.72 MB

Xet Pointer Details

( Raw pointer file )

Xet hash:: d9e41a52114440b48fa53d0c877c7141076ab15c0b129753db090edfb959cc16
Size of remote file:: 1.72 MB
SHA256:: 494837f98ff3bd3ec58b8d24ce3a6b5ef77381b05952e7a064ca5e247115d2c3

Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.