The Llama 3 Herd of Models

TL;DR

This paper introduces Llama 3, a new family of multilingual foundation models with up to 405B parameters, supporting coding, reasoning, and tool use, and evaluates their performance across diverse tasks.

Contribution

It presents the development and extensive evaluation of Llama 3, including multimodal extensions, and releases the models for public use with safety measures.

Findings

Llama 3 achieves comparable quality to GPT-4 on many tasks.

Multimodal extensions perform competitively on image, video, and speech recognition.

Models are not yet broadly released due to ongoing development.

Abstract

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.