Instructions to use kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit with Transformers:

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

pipe = pipeline("text-generation", model="kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit")
model = AutoModelForMultimodalLM.from_pretrained("kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit

SGLang

How to use kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-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 "kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit with Docker Model Runner:
```
docker model run hf.co/kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit
```

Model Details

This is Qwen/Qwen2.5-1.5B quantized with AutoRound (asymmetric quantization) and serialized with the GPTQ format in 4-bit. The model has been created, tested, and evaluated by The Kaitchup.

Details on the quantization process and how to use the model here: The Best Quantization Methods to Run Llama 3.1 on Your GPU

It is possible to fine-tune an adapter on top of it following the QLoRA methodology. More about this here: QLoRA with AutoRound: Cheaper and Better LLM Fine-tuning on Your GPU

I used these hyperparameters for quantization:

bits, group_size = 4, 128

autoround = AutoRound(model, tokenizer, nsamples=512, iters=1000, low_gpu_mem_usage=False, bits=bits, group_size=group_size)

autoround.quantize()
output_dir = "./tmp_autoround"
autoround.save_quantized(output_dir, format='auto_gptq', inplace=True)

Evaluation results (zero-shot evaluation with lm_eval):

Developed by: The Kaitchup
Language(s) (NLP): English
License: Apache 2.0 license

Downloads last month: 5

Safetensors

Model size

2B params

Tensor type

I32

F16

Collection including kaitchup/Qwen2.5-1.5B-AutoRound-GPTQ-asym-4bit

Quantized Qwen2.5

Collection

9 items • Updated Dec 9, 2024 • 4