Instructions to use HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4 with Transformers:

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

pipe = pipeline("text-generation", model="HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4")
model = AutoModelForCausalLM.from_pretrained("HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4

SGLang

How to use HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4 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 "HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4" \
    --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": "HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4",
		"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 "HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4" \
        --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": "HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4 with Docker Model Runner:
```
docker model run hf.co/HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4
```

WarpQuant Llama 3 8B R16E4H4

Llama 3 8B quantized with signed Hadamard rotation, block-GPTQ, and Output-Fisher weak-column recovery. Projection weights use a 3.5-bpw INT3 base, selected columns are restored in BF16, and the embedding and output head use group-128 INT4.

Technical report · GitHub

Payload and evaluation

Format	Text bpw	Payload	WikiText-2 PPL ↓	ARC-299 ↑	MMLU-13,943 ↑
BF16	16.00	14.965 GiB	6.2559	50.50	41.04
Q4_K_M	4.89	4.583 GiB	6.4359	50.84	40.67
IQ3_S + imatrix	3.66	3.429 GiB	6.9929	44.15	39.87
WarpQuant Fisher R16E4H4	3.6256	3.389 GiB	7.3446	45.49	38.99

The repository stores the quantized values in BF16-compatible safetensors. The reported payload is the packed-equivalent analytical size including codes, scales, recovery values, and column indices.

Usage

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "HarimxChoi/WarpQuant-Llama-3-8B-R16E4H4"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

Citation

@misc{choi2026warpquant,
  author = {Harim Choi},
  title = {WarpQuant: Dual-Domain LLM Quantization via Hadamard Rotation and Output-Fisher Sensitivity},
  year = {2026},
  url = {https://harimxchoi.github.io/projects/warpquant/}
}