Instructions to use Compumacy/Llama-3.2-3B-Instruct-WINT4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Compumacy/Llama-3.2-3B-Instruct-WINT4 with Transformers:

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

pipe = pipeline("text-generation", model="Compumacy/Llama-3.2-3B-Instruct-WINT4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Compumacy/Llama-3.2-3B-Instruct-WINT4")
model = AutoModelForCausalLM.from_pretrained("Compumacy/Llama-3.2-3B-Instruct-WINT4")
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

vLLM

How to use Compumacy/Llama-3.2-3B-Instruct-WINT4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Compumacy/Llama-3.2-3B-Instruct-WINT4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Compumacy/Llama-3.2-3B-Instruct-WINT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Compumacy/Llama-3.2-3B-Instruct-WINT4

SGLang

How to use Compumacy/Llama-3.2-3B-Instruct-WINT4 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 "Compumacy/Llama-3.2-3B-Instruct-WINT4" \
    --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": "Compumacy/Llama-3.2-3B-Instruct-WINT4",
		"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 "Compumacy/Llama-3.2-3B-Instruct-WINT4" \
        --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": "Compumacy/Llama-3.2-3B-Instruct-WINT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Compumacy/Llama-3.2-3B-Instruct-WINT4 with Docker Model Runner:
```
docker model run hf.co/Compumacy/Llama-3.2-3B-Instruct-WINT4
```

Llama-3.2-3B-Instruct-WINT4

This model is a 4-bit quantized version of meta-llama/Llama-3.2-3B-Instruct "using the llmcompressor library.

Quantization Details

Base Model: meta-llama/Llama-3.2-3B-Instruct
Quantization Library: llmcompressor
Quantization Method: Weight-only 4-bit int (WINT4)
Quantization Recipe:

quant_stage:
      quant_modifiers:
        QuantizationModifier:
          ignore: [lm_head]
          config_groups:
            group_0:
              weights: {num_bits: 4, type: int, symmetric: true, strategy: channel, dynamic: false}
              targets: [Linear]

Evaluation Results

The following table shows the evaluation results on various benchmarks compared to the baseline (non-quantized) model.

Task	Baseline Metric (10.0% Threshold)	Quantized Metric	Metric Type
winogrande	0.7119	0.6606	acc,none

How to Use

You can load the quantized model and tokenizer using the transformers library:

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "NoorNizar/Llama-3.2-3B-Instruct-WINT4"

model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(model_id)

# Example usage (replace with your specific task)
prompt = "Hello, world!"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=50)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Disclaimer

This model was quantized automatically using a script. Performance and behavior might differ slightly from the original base model.

Downloads last month: 2

Safetensors

Model size

4B params

Tensor type

F16