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

Libraries

How to use iproskurina/gemma-2b-gptqmodel-4bit with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

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

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use iproskurina/gemma-2b-gptqmodel-4bit with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/iproskurina/gemma-2b-gptqmodel-4bit

SGLang

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

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

A newer version of this model is available: google/gemma-2-2b

gemma-2b - GPTQ (4-bit)

Source model: google/gemma-2b

This model was quantized to 4-bit using GPTQModel.

Quantization parameters:

bits: 4
group_size: 128
damp_percent: 0.05
desc_act: False

Usage

            # pip install transformers gptqmodel --no-build-isolation
            from gptqmodel import GPTQModel
            model_id = "iproskurina/gemma-2b-gptqmodel-4bit"
            model = GPTQModel.load(model_id)
            result = model.generate("Uncovering deep insights")[0]
            print(model.tokenizer.decode(result))

Downloads last month: -

Safetensors

Model size

3B params

Tensor type

BF16

I32

F32

iproskurina
/

gemma-2b-gptqmodel-4bit

gemma-2b - GPTQ (4-bit)

Usage

Dataset used to train iproskurina/gemma-2b-gptqmodel-4bit