Instructions to use amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2 with Transformers:

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

pipe = pipeline("text-generation", model="amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2")
model = AutoModelForCausalLM.from_pretrained("amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2", device_map="auto")
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 amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2

SGLang

How to use amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2 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 "amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2" \
    --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": "amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2",
		"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 "amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2" \
        --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": "amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2 with Docker Model Runner:
```
docker model run hf.co/amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2
```

Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2

Model Overview

Model Architecture: LlamaForCausalLM
- Input: Text
- Output: Text
Source Model: Llama-3.1-8B-Instruct
Supported Hardware: AMD EPYC (CPU inference)
Preferred Operating System: Linux
Inference Engine: vLLM v0.25.1
Quantization Framework: LLM Compressor v0.10.0.2
Quantization Method: 4-bit Weight-Only Quantization (W4A16)
Compatible Stack:
- ZenDNN v6.1.0
- ZenTorch v2.11.0.3
- PyTorch v2.11.0
- LLM Compressor v0.10.0.2
- vLLM v0.25.1

This model was Built with Llama. This is a quantized version of Llama-3.1-8B-Instruct created by AMD using LLM Compressor (compressed-tensors) for ZenDNN-optimized CPU inference.

Quantization

The model was quantized from Llama-3.1-8B-Instruct using LLM Compressor.

Method: 4-bit Weight-Only Quantization (W4A16)
Config: compressed-tensors, num_bits=4, type=int, symmetric=true, group_size=128
Weights: INT4 (4-bit integer, symmetric)
Activations: FP16 (16-bit, unquantized)
Group Size: 128

Quantization script:

python llmcompressor_w4a16.py \
    --model_name meta-llama/Llama-3.1-8B-Instruct \
    --output_dir ./Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2

Quick Start

Use with vLLM

from vllm import LLM, SamplingParams

model = LLM(
    model="amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2",
    dtype="bfloat16",
)

sampling_params = SamplingParams(temperature=0.7, max_tokens=256)
outputs = model.generate(["Hello, how are you?"], sampling_params)
print(outputs[0].outputs[0].text)

Requirements

torch==2.11.0
zentorch==2.11.0.3
vllm==0.25.1

OpenMP Setup

For optimal performance, set LD_PRELOAD with libomp.so (LLVM OpenMP) or libiomp5.so (Intel OpenMP):

# Using LLVM OpenMP (llvmopenmp)
export LD_PRELOAD=$(find /path/to/env -name "libomp.so" | head -1)

# Or using Intel OpenMP (libiomp)
export LD_PRELOAD=$(find /path/to/env -name "libiomp5.so" | head -1)

Note: Set LD_PRELOAD before launching vLLM or any inference script.

Evaluation

The model was evaluated against the BF16 (unquantized) baseline on standard benchmarks using lm-evaluation-harness with the vLLM engine.

Benchmark	BF16 Baseline	W4A16 (this model)	Recovery
MMLU (5-shot)	-	-	-
GSM8K_COT (8-shot)	-	-	-
Perplexity (wikitext2)	-	-	-

Evaluation results will be updated after benchmarking.

Reproduction

lm_eval \
    --model vllm \
    --model_args pretrained="amd/Llama-3.1-8B-Instruct-w4a16-llmcompressor-v0.10.0.2" \
    --tasks mmlu \
    --num_fewshot 5 \
    --batch_size auto

Limitations

Version Lock: This model is compatible with ZenDNN v6.1.0 / PyTorch v2.11.0. It may not load correctly on other versions.
CPU Only: This model is optimized for AMD EPYC CPU inference via ZenDNN. It is not intended for GPU inference.