Instructions to use yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit with Transformers:

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

pipe = pipeline("text-generation", model="yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit")
model = AutoModelForCausalLM.from_pretrained("yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit")
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 yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit

SGLang

How to use yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit 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 "yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit" \
    --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": "yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit",
		"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 "yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit" \
        --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": "yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit with Docker Model Runner:
```
docker model run hf.co/yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit
```

Meta-Llama-3.1-8B-Instruct AWQ 2-bit

This repository contains an AWQ-style weight-only quantized derivative of meta-llama/Meta-Llama-3.1-8B-Instruct.

Quantization

Method: AWQ
Weight bits: 2
Activation precision: 16-bit
Scheme: W2A16 asymmetric group quantization
Group size: 128
Calibration dataset: c4
Backend path: Hugging Face transformers + llm-compressor
Experimental low-bit AWQ: yes

Validation

Validation has not been run yet.

License and Use

This model is derived from Meta Llama 3.1 materials. Use and redistribution must comply with the Llama 3.1 Community License, the Acceptable Use Policy, and any Hugging Face gated-model terms for the base checkpoint.

Run Metadata

{
  "activation_dtype": "float16_or_bfloat16",
  "base_model": "meta-llama/Meta-Llama-3.1-8B-Instruct",
  "bits": 2,
  "config": "/nas/longleaf/home/yuanwu/Bias_Compressed_LLM/Quantization/AWQ/configs/llama31_8b_instruct.yaml",
  "created_at": "2026-06-22T07:04:16.871307+00:00",
  "elapsed_seconds": 679.45,
  "environment": {
    "cuda_available": true,
    "cuda_devices": [
      {
        "capability": "8.9",
        "index": 0,
        "name": "NVIDIA L40S",
        "total_memory_gb": 44.39
      }
    ],
    "platform": "Linux-5.14.0-611.16.1.el9_7.x86_64-x86_64-with-glibc2.34",
    "python": "3.10.20",
    "torch": "2.11.0+cu130",
    "torch_cuda": "13.0"
  },
  "experimental_low_bit_awq": true,
  "method": "AWQ",
  "output_dir": "/users/y/u/yuanwu/Bias_Compressed_LLM/awq_outputs/Meta-Llama-3.1-8B-Instruct-AWQ-2bit",
  "scheme": "W2A16"
}

Downloads last month: 37

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for yw223/Meta-Llama-3.1-8B-Instruct-AWQ-2bit

Base model

meta-llama/Llama-3.1-8B

Finetuned

meta-llama/Llama-3.1-8B-Instruct

Quantized

(886)

this model