Instructions to use daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1 with Transformers:

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

pipe = pipeline("text-generation", model="daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1

SGLang

How to use daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1 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 "daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1" \
    --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": "daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1",
		"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 "daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1" \
        --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": "daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1 with Docker Model Runner:
```
docker model run hf.co/daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1
```

Llama-3.1-5B — 40% Compressed from Llama-3.1-8B (English)

Built with Llama. This repository is part of the Efficient and Robust AI System (E-AI) Project by Vincent-Daniel Yun. This model is a compressed edition of meta-llama/Llama-3.1-8B with 13 of 32 transformer layers removed (19 layers remain, ≈5.19B parameters), recovered training-free. It is a base (non–chat-tuned) model.

🔗 Project: https://www.worldwidedaniel.com/eai-project 📅 Release: 2026-06-28 · Version: V1

⚖️ License: governed by the Llama 3.1 Community License Agreement (https://huggingface.co/meta-llama/Llama-3.1-8B/blob/main/LICENSE), and subject to the Llama 3.1 Acceptable Use Policy (https://www.llama.com/llama3_1/use-policy). By using this model you agree to those terms. "Llama" is a trademark of Meta Platforms, Inc. Built with Llama.

⚠️ Language: English-focused. Use as a discrimination / classification engine (classification, reading comprehension, domain QA, judging, safety screening) — open-ended long-form generation is degraded by compression (see PPL). For factual questions use retrieval (RAG).

About E-AI

Modern AI is powerful but heavy. State-of-the-art models are enormous and their inference is slow. The E-AI (Efficient-AI) project builds compact yet capable AI — making every model lightweight and fast, and keeping multi-agent teams reliable even when some members fail — so AI can assist people in urgent, high-stakes moments.

Method

The pruning method and the training-free recovery method are proprietary, undisclosed methods created by Vincent-Daniel Yun and are not released. Only the resulting model is shared.

Results (measured, full test sets)

PPL on 2048-token context (↓ better); downstream & MMLU are 0-shot via lm-eval-harness (↑ better).

Metric	Llama-3.1-8B (base)	This model (40%)
PPL · WikiText2 ↓	6.24	434.58
PPL · C4 ↓	8.68	260.54
ARC-c ↑	0.5350	0.3686
ARC-e ↑	0.8114	0.5105
BoolQ ↑	0.8208	0.6223
COPA ↑	0.8700	0.7300
HellaSwag ↑	0.7889	0.4771
OpenBookQA ↑	0.4480	0.3180
RACE ↑	0.3914	0.3244
RTE ↑	0.6931	0.6173
WinoGrande ↑	0.7356	0.6283
Avg. downstream (9) ↑	0.6771	0.5107
MMLU ↑	0.6360	0.5598

Task-suitability (vs base Llama-3.1-8B)

Where this compressed model stays close to the base on discrimination tasks (full test sets):

Task	Llama-3.1-8B (base)	This model (40%)
Topic classification (AG News)	0.564	0.356
LLM-as-judge (RewardBench)	0.660	0.688
SafetyBench (MCQ)	0.743	0.629
MultiRC	0.572	0.572
WiC	0.511	0.500
MRPC	0.674	0.684
CB (NLI)	0.643	0.518
SST-2	0.775	0.509
MedQA	0.598	0.577
MedMCQA	0.564	0.530
PubMedQA	0.758	0.562
Belebele-en	0.788	0.629
Belebele-ko	0.662	0.543
XNLI-zh	0.395	0.338
ToxiGen	0.432	0.430
TruthfulQA	0.452	0.501

(Discrimination is largely preserved; generation/PPL degrades with compression. Yes/No safety-F1 benchmarks are omitted as the non-chat base is poorly calibrated for them.)

Efficiency (measured, fp16, single GPU)

	Llama-3.1-8B (base)	This model (40%)
Layers	32	19
Parameters	8.03B	5.19B
Peak inference memory (fp16)	19.31 GB	13.2 GB (−32%)
Forward latency (fp16)	1293 ms	812 ms (−37%)

Usage — Transformers

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
m = AutoModelForCausalLM.from_pretrained("daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1", trust_remote_code=True, dtype=torch.float16, device_map="cuda")
tok = AutoTokenizer.from_pretrained("daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1", trust_remote_code=True)
ids = tok("The capital of France is", return_tensors="pt").to("cuda")
print(tok.decode(m.generate(**ids, max_new_tokens=20)[0]))

trust_remote_code=True is required (custom decoder layer in modeling_llama_recovered.py).

License & Acknowledgements

This model is a derivative of Llama 3.1 and is distributed under the Llama 3.1 Community License. Built with Llama. Thanks to Prof. Sai Praneeth Karimireddy (USC) and Prof. Sunwoo Lee (Inha University) for guidance, and to Meta for releasing Llama 3.1.

Downloads last month: -

Safetensors

Model size

5B params

Tensor type

F16

Model tree for daniel-eai/Llama-3.1-5B-40pct-Compressed-8B-EN-V1

Base model

meta-llama/Llama-3.1-8B

Finetuned

(1435)

this model