---
license: llama3.1
language:
- en
- de
- fr
- it
- pt
- hi
- es
- th
library_name: transformers
pipeline_tag: text-generation
tags:
- llama-3.1
- meta
- bnb
---

> [!IMPORTANT]
> This repository is a community-driven quantized version of the original model [`meta-llama/Meta-Llama-3.1-405B-Instruct`](https://huggingface.co/meta-llama/Meta-Llama-3.1-405B-Instruct) which is the BF16 half-precision official version released by Meta AI.

## 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 contains [`meta-llama/Meta-Llama-3.1-405B-Instruct`](https://huggingface.co/meta-llama/Meta-Llama-3.1-405B-Instruct) quantized using [bitsandbytes](https://github.com/bitsandbytes-foundation/bitsandbytes) from BF16 down to NF4 with a block size of 64.

## Model Usage

> [!NOTE]
> In order to run the inference with Llama 3.1 405B Instruct BNB in NF4, around 220 GiB of VRAM are needed only for loading the model checkpoint, without including the KV cache or the CUDA graphs, meaning that there should be a bit over that VRAM available.

In order to use the current quantized model, support is offered for different solutions as `transformers`, or `text-generation-inference`.

### 🤗 transformers

In order to run the inference with Llama 3.1 405B Instruct BNB in NF4, both `torch` and `bitsandbytes` need to be installed as:

```bash
pip install "torch>=2.0.0" bitsandbytes --upgrade
```

Then, the latest version of `transformers` need to be installed, being 4.43.0 or higher, as:

```bash
pip install "transformers[accelerate]>=4.43.0" --upgrade
```

To run the inference on top of Llama 3.1 405B Instruct BNB in NF4 precision, the model can be instantiated as any other causal language modeling model via `AutoModelForCausalLM` and run the inference normally.

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "hugging-quants/Meta-Llama-3.1-405B-Instruct-BNB-NF4"
prompt = [
  {"role": "system", "content": "You are a helpful assistant, that responds as a pirate."},
  {"role": "user", "content": "What's Deep Learning?"},
]

tokenizer = AutoTokenizer.from_pretrained(model_id)

inputs = tokenizer.apply_chat_template(prompt, tokenize=True, add_generation_prompt=True, return_tensors="pt").cuda()

model = AutoModelForCausalLM.from_pretrained(
  model_id,
  torch_dtype=torch.bfloat16,
  low_cpu_mem_usage=True,
  device_map="auto",
)

outputs = model.generate(inputs, do_sample=True, max_new_tokens=256)
print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
```