Companies from corner stores to retail giants are realizing that pocket-sized artificial intelligence (AI) can lead to significant cost savings. This is because agile new language models can compete at a fraction of the cost of large legacy models.
H2O.ai recently announced a smaller model as an example of a broader shift towards compact AI models. It promises to reshape the way businesses handle everything from inventory to customer service. These streamlined systems can process purchase orders, receipts, and shopping data with the same accuracy as large language models while reducing computing costs.
“Small AI models can level the playing field between large retailers and small and medium-sized businesses (SMBs) by providing cost-effective solutions for inventory management and customer service automation. ” payments company Radial told PYMNTS. “These models allow small and medium-sized businesses to optimize inventory levels, predict demand, and automate reordering while improving operational efficiency.”
Small-scale language models are streamlined AI systems that process text using millions rather than billions of parameters. Get faster performance and lighter computing requirements in exchange for some advanced features.
In AI, small becomes the new big
H2O.ai has released two small AI models (800 million parameters and 2 billion parameters) for document processing and text recognition. The model is trained on millions of conversation pairs and is available for free on Hugging Face. The company claims that this 800 million model outperforms major competitors on optical character recognition (OCR) benchmarks.
“We designed the H2OVL Mississippi model to be a high-performance, yet cost-effective solution to bring AI-powered OCR, visual understanding, and document AI to your business,” CEO and Founder of H2O.ai said Sri Ambati. News release. “By combining cutting-edge multimodal AI with highly efficient capabilities, H2OVL Mississippi delivers accurate and scalable document AI solutions to a wide range of industries.”
Sermarini pointed out that smaller models like the H20 are typically faster to prototype and develop.
“This will enable new and creative players to quickly design advanced analytics solutions for retail, supply chain and logistics,” he said.
Stephen DeAngelis, founder and CEO of AI company Enterra Solutions, says smaller, more efficient models reduce computing barriers, but bringing AI to the logistics and retail sector requires more than just cost savings. told PYMNTS.
“Language models still require extensive training, high-quality data, and expertise to tackle the operational challenges of complex supply chain management and business analysis,” he added. “Technological transition is likely to be gradual rather than immediate.”
The rise of efficient and compact AI models is opening the door to startups that can’t afford huge computing costs. These lightweight models allow startups to compete and innovate without requiring huge budgets for AI infrastructure, Hardik Chawla, Amazon’s senior product manager for technology, told PYMNTS. He said he is seeing more companies building robust retail solutions that run on smaller hardware.
“Instead of requiring large GPU clusters, startups can deploy focused models that handle specific high-value tasks, from dynamic pricing to demand forecasting,” he said. I added.
“Consider a startup focused on predicting out-of-stocks. Rather than building a comprehensive retail AI system, it deploys small, efficient models trained specifically on inventory patterns. This focused approach not only reduces development and operational costs, but also often yields better results than trying to extract this functionality from a large, generic model.
help rather than replace humans
Despite the benefits of small-scale AI models, observers do not expect AI to replace human workers anytime soon. Chawla said retail, warehousing and customer service could potentially automate many routine manual tasks, reducing time spent on repetitive tasks.
For example, in customer service, Chawla said AI could handle common inquiries such as order status and allow agents to provide more personalized support. This enhanced approach increases efficiency, improves quality of service, and makes personalized, AI-assisted customer interactions scalable to businesses of all sizes.
Similarly, in warehouse operations, these models can assist workers by optimizing workflows, suggesting the most efficient routes or flagging potential inventory issues in real time. can. Workers will still handle physical tasks, but AI will support them by reducing errors and increasing productivity.
“Regardless of the industry, data literacy, AI monitoring, and technology-driven process management will be highly sought-after skills,” Chawla said. “Avoiding displacement will require a major push towards reskilling and upskilling efforts – not eliminating jobs entirely, but shifting the workforce into new types of roles.”
See more: AI, Amazon, Artificial Intelligence, Customer Service, Enterra Solutions, GenAI, Generated AI, H2O.ai, Hardik Chawla, Hugging Face, Inventory Management, News, PYMNTS News, Radial, Small AI, Small Business, Small Language Model, SMB, Shri Ambati, Stephen DeAngelis, Stephen Sermarini
Source link
