Command A: An Enterprise-Ready Large Language Model

In this report we describe the development of Command A, a powerful large language model purpose-built to excel at real-world enterprise use cases. Command A is an agent-optimised and multilingual-capable model, with support for 23 languages of global business, and a novel hybrid architecture balancing efficiency with top of the range performance. It offers best-in-class Retrieval Augmented Generation (RAG) capabilities with grounding and tool use to automate sophisticated business processes. These abilities are achieved through a decentralised training approach, including self-refinement algorithms and model merging techniques. We also include results for Command R7B which shares capability and architectural similarities to Command A. Weights for both models have been released for research purposes. This technical report details our original training pipeline and presents an extensive evaluation of our models across a suite of enterprise-relevant tasks and public benchmarks, demonstrating excellent performance and efficiency.

Introduction. Large Language Models (LLMs) are Artificial Intelligence (AI) models designed to understand and generate human-like text conditioned on the input they receive. Recent advancements have led to remarkable breakthroughs in their ability to comprehend and produce human language with unparalleled accuracy and fluency. This progress has been instrumental in their widespread adoption across various real-world and enterprise environments, where they significantly boost operational efficiency and deepen understanding. This technical report describes the development of Command A and Command R7B, two LLMs designed to excel in real-world enterprise settings. Both the 111B parameter Command A and Command R7B perform best-in-class across a suite of established benchmarks for their respective model sizes.

Discussion / Conclusion. This technical report detailed the development of Command A, shared extensive performance evaluations across many domains and languages, and shared additional results for Command R7B. Command A represents a significant advancement in LLMs for enterprise, achieving best-in-class performance across a wide range of tasks with optimal efficiency. Our models excel in enterprise-relevant tasks such as agentic workflows, multilingual understanding and generation, and instruction-following. Key innovations introduced include data and architectural optimisations, self-refinement algorithms, and a model merging-based approach that ensures expert-level performance across diverse capabilities within a single model. Command A outperforms comparable models in both efficiency and computational overhead, requiring fewer resources for serving, making it easy to deploy on-premises or in private cloud environments on just two A100 or H100 GPUs, and delivering tokens at a higher rate. The release of model weights under a non-commercial license further facilitates community-based exploration and research. Command A sets a new standard for LLMs in enterprise applications, balancing performance, efficiency, and versatility — and providing maximum performance for minimal compute.