Instructions to use iproskurina/opt-2.7b-GPTQ-4bit-g128 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use iproskurina/opt-2.7b-GPTQ-4bit-g128 with Transformers:

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

pipe = pipeline("text-generation", model="iproskurina/opt-2.7b-GPTQ-4bit-g128")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("iproskurina/opt-2.7b-GPTQ-4bit-g128")
model = AutoModelForCausalLM.from_pretrained("iproskurina/opt-2.7b-GPTQ-4bit-g128", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use iproskurina/opt-2.7b-GPTQ-4bit-g128 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "iproskurina/opt-2.7b-GPTQ-4bit-g128"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "iproskurina/opt-2.7b-GPTQ-4bit-g128",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/iproskurina/opt-2.7b-GPTQ-4bit-g128

SGLang

How to use iproskurina/opt-2.7b-GPTQ-4bit-g128 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/opt-2.7b-GPTQ-4bit-g128" \
    --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/opt-2.7b-GPTQ-4bit-g128",
		"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/opt-2.7b-GPTQ-4bit-g128" \
        --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/opt-2.7b-GPTQ-4bit-g128",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use iproskurina/opt-2.7b-GPTQ-4bit-g128 with Docker Model Runner:
```
docker model run hf.co/iproskurina/opt-2.7b-GPTQ-4bit-g128
```

opt-2.7b-GPTQ-4bit-g128 / README.md

iproskurina

Update README.md to include GPTQModel usage.

2b9fd2a verified over 1 year ago

preview code

Raw History Blame Contribute Delete

2.75 kB

	---
	base_model: facebook/opt-2.7b
	language:
	- en
	license: other
	model_name: opt-2.7b
	pipeline_tag: text-generation
	inference: false
	model_creator: facebook
	model_type: opt
	quantized_by: iproskurina
	tags:
	- gptq
	- 4-bit
	base_model_relation: quantized
	---


	<img src="https://cdn-uploads.huggingface.co/production/uploads/629a3dbcd496c6dcdebf41cc/t-6kpqFpEYJPT6zmvnm49.png" width="200" />

	# OPT-2.7B - GPTQ


	- Model creator: [Meta AI](https://huggingface.co/facebook)
	- Original model: [OPT-2.7B](https://huggingface.co/facebook/opt-2.7b)

	The model published in this repo was quantized to 4bit using [AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ).

	Quantization details

	All quantization parameters were taken from [GPTQ paper](https://arxiv.org/abs/2210.17323).

	GPTQ calibration data consisted of 128 random 2048 token segments from the [C4 dataset](https://huggingface.co/datasets/c4).

	The grouping size used for quantization is equal to 128.

	## How to use this GPTQ model from Python code

	### Install the necessary packages

	Requires: Transformers 4.33.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.

	```shell
	pip3 install --upgrade transformers optimum
	# If using PyTorch 2.1 + CUDA 12.x:
	pip3 install --upgrade auto-gptq
	# or, if using PyTorch 2.1 + CUDA 11.x:
	pip3 install --upgrade auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/
	```

	If you are using PyTorch 2.0, you will need to install AutoGPTQ from source. Likewise if you have problems with the pre-built wheels, you should try building from source:

	```shell
	pip3 uninstall -y auto-gptq
	git clone https://github.com/PanQiWei/AutoGPTQ
	cd AutoGPTQ
	git checkout v0.5.1
	pip3 install .
	```

	### You can then use the following code

	```python

	from transformers import AutoTokenizer, TextGenerationPipeline,AutoModelForCausalLM
	from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig
	pretrained_model_dir = "iproskurina/opt-2.7b-gptq-4bit"
	tokenizer = AutoTokenizer.from_pretrained(pretrained_model_dir, use_fast=True)
	model = AutoGPTQForCausalLM.from_quantized(pretrained_model_dir, device="cuda:0", model_basename="model")
	pipeline = TextGenerationPipeline(model=model, tokenizer=tokenizer)
	print(pipeline("auto-gptq is")[0]["generated_text"])
	```

	[LICENSE](https://huggingface.co/facebook/opt-2.7b/blob/main/LICENSE.md)
	### Run the model with GPTQModel

	GPTQModel package: https://github.com/ModelCloud/GPTQModel

	```
	pip install -v gptqmodel=="1.8.0" --no-build-isolation
	from gptqmodel import GPTQModel

	model_id = 'iproskurina/opt-2.7b-GPTQ-4bit-g128'
	model = GPTQModel.load(model_id)
	result = model.generate("Uncovering deep insights")[0] # tokens
	print(model.tokenizer.decode(result)) # string output
	```