Toronto-based Cohere has launched Command A, a new generative AI model designed to deliver high performance with lower processing power. The model runs on just two Nvidia A100 or H100 chips, compared to the dozens required by competitors, making it cost-effective for enterprise adoption.
Cohere claims Command A rivals OpenAI’s GPT-4o and DeepSeek V3 in performance, excelling in academic tasks, retail workflows, and code generation. Integrated into Cohere’s North platform, the model enhances AI agents for HR, finance, and enterprise applications. Pricing remains in line with OpenAI’s GPT-4o.
Want to know more? Check out the source code on The Logic.