Granite Code Models: A Family of Open Foundation Models for Code Intelligence
Abstract
Large Language Models (LLMs) trained on code are revolutionizing the software development process. Increasingly, code LLMs are being integrated into software development environments to improve the productivity of human programmers, and LLM-based agents are beginning to show promise for handling complex tasks autonomously. Realizing the full potential of code LLMs requires a wide range of capabilities, including code generation, fixing bugs, explaining and documenting code, maintaining repositories, and more. In this work, we introduce the Granite series of decoder-only code models for code generative tasks, trained with code written in 116 programming languages. The Granite Code models family consists of models ranging in size from 3 to 34 billion parameters, suitable for applications ranging from complex application modernization tasks to on-device memory-constrained use cases. Evaluation on a comprehensive set of tasks demonstrates that Granite Code models consistently reaches state-of-the-art performance among available open-source code LLMs. The Granite Code model family was optimized for enterprise software development workflows and performs well across a range of coding tasks (e.g. code generation, fixing and explanation), making it a versatile all around code model. We release all our Granite Code models under an Apache 2.0 license for both research and commercial use.
Cite
@article{arxiv.2405.04324,
title = {Granite Code Models: A Family of Open Foundation Models for Code Intelligence},
author = {Mayank Mishra and Matt Stallone and Gaoyuan Zhang and Yikang Shen and Aditya Prasad and Adriana Meza Soria and Michele Merler and Parameswaran Selvam and Saptha Surendran and Shivdeep Singh and Manish Sethi and Xuan-Hong Dang and Pengyuan Li and Kun-Lung Wu and Syed Zawad and Andrew Coleman and Matthew White and Mark Lewis and Raju Pavuluri and Yan Koyfman and Boris Lublinsky and Maximilien de Bayser and Ibrahim Abdelaziz and Kinjal Basu and Mayank Agarwal and Yi Zhou and Chris Johnson and Aanchal Goyal and Hima Patel and Yousaf Shah and Petros Zerfos and Heiko Ludwig and Asim Munawar and Maxwell Crouse and Pavan Kapanipathi and Shweta Salaria and Bob Calio and Sophia Wen and Seetharami Seelam and Brian Belgodere and Carlos Fonseca and Amith Singhee and Nirmit Desai and David D. Cox and Ruchir Puri and Rameswar Panda},
journal= {arXiv preprint arXiv:2405.04324},
year = {2024}
}
Comments
Corresponding Authors: Rameswar Panda, Ruchir Puri; Equal Contributors: Mayank Mishra, Matt Stallone, Gaoyuan Zhang