Instructions to use kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8 with Transformers:

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

pipe = pipeline("text-generation", model="kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8")
model = AutoModelForCausalLM.from_pretrained("kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8

SGLang

How to use kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8 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 "kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8" \
    --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": "kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8",
		"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 "kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8" \
        --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": "kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8 with Docker Model Runner:
```
docker model run hf.co/kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8
```

Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8

This is Llama 3.2 3B quantized to 2 bits per parameter using AQLM with OA-EM initialization and PV-tuning.

OA-EM (Output-Aware Expectation-Maximisation) is a Hessian-weighted EM algorithm that significantly improves codebook initialization for additive quantization, particularly at extreme compression rates. See the paper Initialisation Determines the Basin: Efficient Codebook Optimisation for Extreme LLM Quantization for details.

Quantization Details

Parameter	Value
Method	AQLM + OA-EM init + PV-tuning
Bitrate	2 bpp (2 codebooks × 8-bit, group size 8)
Codebook config	`num_codebooks=2, nbits_per_codebook=8, in_group_size=8`
Beam size	4
OA-EM config	3 rounds, 100 Adam steps, lr=1e-4
PV-tuning	5 epochs, LAMB optimizer, lr=3e-4, 10K C4 samples
Seed	42
Context length	4096 tokens

Results

Perplexity (WikiText-2)

Model	Pre-PV	Post-PV
FP16 (unquantized)	7.28	—
AQLM Greedy (b=4)	352.39	12.66
AQLM OA-EM (b=4)	16.82	11.53

Downstream Tasks (post-PV-tuning, b=4)

Task	Metric	AQLM Greedy	AQLM OA-EM
ARC-Challenge	acc_norm ↑	.350	.359
ARC-Easy	acc_norm ↑	.560	.614
HellaSwag	acc_norm ↑	.620	.625
LAMBADA	acc ↑	.577	.584
LAMBADA	ppl ↓	7.65	7.13
PIQA	acc_norm ↑	.734	.734
WinoGrande	acc ↑	.594	.619
Average		.573	.589

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8",
    trust_remote_code=True,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8")

inputs = tokenizer("The meaning of life is", return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Note: Requires the aqlm inference library:

pip install aqlm[gpu]

Paper

Initialisation Determines the Basin: Efficient Codebook Optimisation for Extreme LLM Quantization

Ian W. Kennedy and Nafise Sadat Moosavi, University of Sheffield

arXiv:2604.08118 | Code

Citation

@article{kennedy2026oaem,
  title={Initialisation Determines the Basin: Efficient Codebook Optimisation for Extreme LLM Quantization},
  author={Kennedy, Ian W. and Moosavi, Nafise Sadat},
  journal={arXiv preprint arXiv:2604.08118},
  year={2026}
}

Downloads last month: 1,533

Model tree for kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8

Base model

meta-llama/Llama-3.2-3B

Finetuned

unsloth/Llama-3.2-3B

Quantized

(54)

this model

Paper for kennedyian94/Llama-3.2-3B-AQLM-OA-EM-2Bit-2x8

Initialisation Determines the Basin: Efficient Codebook Optimisation for Extreme LLM Quantization

Paper • 2604.08118 • Published Apr 9 • 2