EXAONE 3.5: Series of Large Language Models for Real-world Use Cases

TL;DR

EXAONE 3.5 models are instruction-tuned large language models with exceptional real-world instruction following, long-context understanding, and competitive performance across multiple benchmarks, available for research and commercial use.

Contribution

Introduction of the EXAONE 3.5 series with multiple sizes, demonstrating superior benchmark performance and open availability for research and commercial applications.

Findings

Achieved top scores in seven instruction-following benchmarks.

Excelled in four long-context comprehension benchmarks.

Performed competitively across nine general benchmarks.

Abstract

This technical report introduces the EXAONE 3.5 instruction-tuned language models, developed and released by LG AI Research. The EXAONE 3.5 language models are offered in three configurations: 32B, 7.8B, and 2.4B. These models feature several standout capabilities: 1) exceptional instruction following capabilities in real-world scenarios, achieving the highest scores across seven benchmarks, 2) outstanding long-context comprehension, attaining the top performance in four benchmarks, and 3) competitive results compared to state-of-the-art open models of similar sizes across nine general benchmarks. The EXAONE 3.5 language models are open to anyone for research purposes and can be downloaded from https://huggingface.co/LGAI-EXAONE. For commercial use, please reach out to the official contact point of LG AI Research: contact_us@lgresearch.ai.