The NVIDIA Llama 3.1 70B Instruct FP8 model is the quantized version of the Meta's Llama 3.1 70B Instruct model, which is an auto-regressive language model that uses an optimized transformer architecture. For more information, please check here. The NVIDIA Llama 3.1 70B Instruct FP8 model is quantized with TensorRT Model Optimizer.

This model is ready for commercial/non-commercial use.

Third-Party Community Consideration

This model is not owned or developed by NVIDIA. This model has been developed and built to a third-party’s requirements for this application and use case; see link to Non-NVIDIA (Meta-Llama-3.1-70B-Instruct) Model Card.

License/Terms of Use:

nvidia-open-model-license

llama3.1

Model Architecture:

Architecture Type: Transformers
Network Architecture: Llama3.1

Input:

Input Type(s): Text
Input Format(s): String
Input Parameters: Sequences
Other Properties Related to Input: Context length up to 128K

Output:

Output Type(s): Text
Output Format: String
Output Parameters: Sequences
Other Properties Related to Output: N/A

Software Integration:

Supported Runtime Engine(s):

Tensor(RT)-LLM
vLLM

Supported Hardware Microarchitecture Compatibility:

NVIDIA Blackwell
NVIDIA Hopper
NVIDIA Lovelace

Preferred Operating System(s):

Linux

Model Version(s):

The model is quantized with nvidia-modelopt v0.15.1

Datasets:

Calibration Dataset: cnn_dailymail
Evaluation Dataset: MMLU

Inference:

Engine: Tensor(RT)-LLM or vLLM
Test Hardware: H100

Post Training Quantization

This model was obtained by quantizing the weights and activations of Meta-Llama-3.1-70B-Instruct to FP8 data type, ready for inference with TensorRT-LLM. Only the weights and activations of the linear operators within transformers blocks are quantized. This optimization reduces the number of bits per parameter from 16 to 8, reducing the disk size and GPU memory requirements by approximately 50%. On H100, we achieved 1.5x speedup.

Usage

Deploy with TensorRT-LLM

To deploy the quantized checkpoint with TensorRT-LLM, follow the sample commands below with the TensorRT-LLM GitHub repo:

Checkpoint convertion:

python examples/llama/convert_checkpoint.py --model_dir Llama-3.1-70B-Instruct-FP8 --output_dir /ckpt --use_fp8

Build engines:

trtllm-build --checkpoint_dir /ckpt --output_dir /engine

Throughputs evaluation:

Please refer to the TensorRT-LLM benchmarking documentation for details.

Evaluation

Precision	MMLU	GSM8K (CoT)	ARC Challenge	IFEVAL	TPS
BF16	83.3	95.3	93.7	92.1	1356.92
FP8	83.2	94.3	93.2	92.2	2040.30

We benchmarked with tensorrt-llm v0.13 on 8 H100 GPUs, using batch size 1024 for the throughputs with in-flight batching enabled. We achieved **~1.5x** speedup with FP8.

Deploy with vLLM

To deploy the quantized checkpoint with vLLM, follow the instructions below:

Install vLLM from directions here.
To use a Model Optimizer PTQ checkpoint with vLLM, quantization=modelopt flag must be passed into the config while initializing the LLM Engine.

Example deployment on an H100:

from vllm import LLM, SamplingParams

model_id = "nvidia/Llama-3.1-70B-Instruct-FP8"
sampling_params = SamplingParams(temperature=0.8, top_p=0.9)

prompts = [
    "Hello, my name is",
    "The president of the United States is",
    "The capital of France is",
    "The future of AI is",
]

llm = LLM(model=model_id, quantization="modelopt")
outputs = llm.generate(prompts, sampling_params)

# Print the outputs.
for output in outputs:
    prompt = output.prompt
    generated_text = output.outputs[0].text
    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")

This model can be deployed with an OpenAI Compatible Server via the vLLM backend. Instructions here.