Instructions to use elysiantech/gemma-2b-gptq-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use elysiantech/gemma-2b-gptq-4bit with Transformers:

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

pipe = pipeline("text-generation", model="elysiantech/gemma-2b-gptq-4bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("elysiantech/gemma-2b-gptq-4bit")
model = AutoModelForCausalLM.from_pretrained("elysiantech/gemma-2b-gptq-4bit")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use elysiantech/gemma-2b-gptq-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "elysiantech/gemma-2b-gptq-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "elysiantech/gemma-2b-gptq-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/elysiantech/gemma-2b-gptq-4bit

SGLang

How to use elysiantech/gemma-2b-gptq-4bit 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 "elysiantech/gemma-2b-gptq-4bit" \
    --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": "elysiantech/gemma-2b-gptq-4bit",
		"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 "elysiantech/gemma-2b-gptq-4bit" \
        --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": "elysiantech/gemma-2b-gptq-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use elysiantech/gemma-2b-gptq-4bit with Docker Model Runner:
```
docker model run hf.co/elysiantech/gemma-2b-gptq-4bit
```

elysiantech/gemma-2b-gptq-4bit

gemma-2b-gptq-4bit is a version of the 2B base model model that was quantized using the GPTQ method developed by Lin et al. (2023).

Please refer to the Original Gemma Model Card for details about the model preparation and training processes.

Dependencies

auto-gptq – AutoGPTQ was used to quantize the phi-3 model.
vllm==0.4.2 – vLLM was used to host models for benchmarking.

Downloads last month: 9

Safetensors

Model size

3B params

Tensor type

I32

F16

Paper for elysiantech/gemma-2b-gptq-4bit

Gradient-Based Post-Training Quantization: Challenging the Status Quo

Paper • 2308.07662 • Published Aug 15, 2023