Instructions to use QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf with Transformers:

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

pipe = pipeline("text-generation", model="QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf")
model = AutoModelForMultimodalLM.from_pretrained("QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf")
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 QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf

SGLang

How to use QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf 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 "QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf" \
    --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": "QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf",
		"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 "QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf" \
        --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": "QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf with Docker Model Runner:
```
docker model run hf.co/QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf
```

📦 Meta-Llama-3-70B-Instruct-4bit-gguf

meta-llama/Meta-Llama-3-70B-Instruct converted to GUFF format

⭐ Star QuantLLM on GitHub

📖 About This Model

This model is meta-llama/Meta-Llama-3-70B-Instruct converted to GUFF format.

Property	Value
Base Model	meta-llama/Meta-Llama-3-70B-Instruct
Format	GUFF
Quantization	None (Full Precision)
License	apache-2.0
Created With	QuantLLM

🚀 Quick Start

With Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model and tokenizer
model = AutoModelForCausalLM.from_pretrained("QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf")
tokenizer = AutoTokenizer.from_pretrained("QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf")

# Generate text
inputs = tokenizer("Once upon a time", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=100, do_sample=True)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

With QuantLLM

from quantllm import TurboModel

# Load with automatic optimization
model = TurboModel.from_pretrained("QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf")

# Generate
response = model.generate("Write a poem about coding")
print(response)

Requirements

pip install transformers torch

📊 Model Details

Property	Value
Original Model	meta-llama/Meta-Llama-3-70B-Instruct
Format	GUFF
Quantization	Full Precision
License	`apache-2.0`
Export Date	2026-04-24
Exported By	QuantLLM v2.0

🚀 Created with QuantLLM

Convert any model to GGUF, ONNX, or MLX in one line!

from quantllm import turbo

# Load any HuggingFace model
model = turbo("meta-llama/Meta-Llama-3-70B-Instruct")

# Export to any format
model.export("guff", quantization="Q4_K_M")

# Push to HuggingFace
model.push("your-repo", format="guff")

📚 Documentation · 🐛 Report Issue · 💡 Request Feature

Downloads last month: 5

Safetensors

Model size

71B params

Tensor type

F32

BF16

Model tree for QuantLLM/Meta-Llama-3-70B-Instruct-4bit-gguf

Base model

meta-llama/Meta-Llama-3-70B

Finetuned

meta-llama/Meta-Llama-3-70B-Instruct

Quantized

(44)

this model