The paper introduces the Granite series of decoder-only code models, designed to support enterprise software development across a wide range of coding tasks. The models, trained on 116 programming languages, range in size from 3 to 34 billion parameters. Evaluation on various benchmarks demonstrates that Granite Code models consistently achieve state-of-the-art performance among available open-source code LLMs. The models are optimized for enterprise software development workflows and perform well across tasks such as code generation, fixing, and explanation. The Granite Code models are released under an Apache 2.0 license for both research and commercial use. The paper details the data collection, model architecture, training methods, and evaluation results, highlighting the models' versatility and robustness in handling diverse coding tasks.The paper introduces the Granite series of decoder-only code models, designed to support enterprise software development across a wide range of coding tasks. The models, trained on 116 programming languages, range in size from 3 to 34 billion parameters. Evaluation on various benchmarks demonstrates that Granite Code models consistently achieve state-of-the-art performance among available open-source code LLMs. The models are optimized for enterprise software development workflows and perform well across tasks such as code generation, fixing, and explanation. The Granite Code models are released under an Apache 2.0 license for both research and commercial use. The paper details the data collection, model architecture, training methods, and evaluation results, highlighting the models' versatility and robustness in handling diverse coding tasks.