Instructions to use Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits with Transformers:

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

pipe = pipeline("text-generation", model="Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits

SGLang

How to use Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits 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 "Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits" \
    --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": "Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits",
		"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 "Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits" \
        --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": "Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits with Docker Model Runner:
```
docker model run hf.co/Volko76/Qwen2.5-0.5B-Instruct-EXL2-4bits
```

Qwen2.5-0.5B-Instruct-EXL2-4bits / README.md

Volko76

Improve language tag (#1)

8ba846b verified over 1 year ago

preview code

Raw History Blame Contribute Delete

715 Bytes

metadata

base_model: Qwen/Qwen2.5-0.5B
language:
  - zho
  - eng
  - fra
  - spa
  - por
  - deu
  - ita
  - rus
  - jpn
  - kor
  - vie
  - tha
  - ara
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct/blob/main/LICENSE
pipeline_tag: text-generation
tags:
  - chat
  - autoquant
  - exl2

Qwen2.5-0.5B-Instruct

Introduction

This model is based on the Qwen2.5-0.5B-Instruct model and is quantized in 4bits in the EXL2 format using the AutoQuant system : https://colab.research.google.com/drive/1b6nqC7UZVt8bx4MksX7s656GXPM-eWw4

You can learn more about the EXL2 format here : https://github.com/turboderp/exllamav2 Feel free to use it as you want