Instructions to use alphrc/Meta-Llama-3.1-405B-4bits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use alphrc/Meta-Llama-3.1-405B-4bits with Transformers:

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

pipe = pipeline("text-generation", model="alphrc/Meta-Llama-3.1-405B-4bits")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("alphrc/Meta-Llama-3.1-405B-4bits")
model = AutoModelForCausalLM.from_pretrained("alphrc/Meta-Llama-3.1-405B-4bits")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use alphrc/Meta-Llama-3.1-405B-4bits with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "alphrc/Meta-Llama-3.1-405B-4bits"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "alphrc/Meta-Llama-3.1-405B-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/alphrc/Meta-Llama-3.1-405B-4bits

SGLang

How to use alphrc/Meta-Llama-3.1-405B-4bits 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 "alphrc/Meta-Llama-3.1-405B-4bits" \
    --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": "alphrc/Meta-Llama-3.1-405B-4bits",
		"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 "alphrc/Meta-Llama-3.1-405B-4bits" \
        --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": "alphrc/Meta-Llama-3.1-405B-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use alphrc/Meta-Llama-3.1-405B-4bits with Docker Model Runner:
```
docker model run hf.co/alphrc/Meta-Llama-3.1-405B-4bits
```

Model Information

The Meta Llama 3.1 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction tuned generative models in 8B, 70B and 405B sizes (text in/text out). The Llama 3.1 instruction tuned text only models (8B, 70B, 405B) are optimized for multilingual dialogue use cases and outperform many of the available open source and closed chat models on common industry benchmarks.

This repository hosts the 4-bit quantized version of meta-llama/Meta-Llama-3.1-405B, which was converted using MLX from FP16. This process reduced the model's size from 820 GB to 231 GB.

Downloads last month: 2

Safetensors

Model size

64B params

Tensor type

F16

U32