Why enterprises should use small language models – Computerworld

Small is beautiful

The all-conquering rise of AI in the enterprise has seen much use of LLMs (large-language models). This week on InfoWorld we wrote about LiteLLM: an open-source gateway for unified LLM access that allows developers to integrate a diverse range of LLM models as if they were calling OpenAI’s API, with support for fallbacks, budgets, rate limits, and real-time monitoring of API calls. 

Good for those organizations wishing to use multiple LLMs in harness. But many InfoWorld readers of the story wanted to understand the utility of small language models. Smart Answers: the child of our very own proprietary LLM, was happy to offer an opinion based on decades of human insights from IT leaders. 

Despite its LLM background, Smart Answers is happy to espouse the virtues of small. SLMs provide agility, cost-effective resources, and rapid prototyping, while also enhancing the security and privacy of organizational data. They also offer better security and customization. In essence, if you want to keep it in house and under control, an SLM might be the answer.