Benete is addressing the growing demands of elder care in nursing homes, at home and in healthcare settings. The company’s BeneCare service uses sensors and smart algorithms to provide insights into the everyday lives of older people so that caregivers can deliver preventative interventions and foster better health outcomes.
THEIR GOAL
Quicker and easier preventative health planning
Benete wanted to make its BeneCare service — which provides elderly client wellbeing data to help detect changes in abilities — more useful for busy caregivers. The team set out to build a secure, multilingual generative AI solution that could convert client data into easy-to-consume summaries. The goal? Streamline everyday care and enable preventative health strategies.
THEIR SOLUTION
An AI-powered summarization tool for fast client care insights
Using IBM watsonx.ai and Llama 3.3 70B Instruct, Benete developed an AI solution that converts client wellbeing data, such as sensor readings, into natural-language text and audio summaries. The solution enabled caregivers to quickly access important information even on the go.The Llama model stood out for its multilingual capabilities and performance, including average generation time. Because Llama is an open-source model, Benete was able to test the new solution without heavy capital investment and save costs while also protecting sensitive client health information.
THEIR APPROACH
Using structured databases and instruction tuning
The solution uses a contextual prompt builder to filter structured JSON data and determine the purpose of a request. The Llama model interprets the data to create summary text, which is sent through to the user interface and stored.To ensure relevance and accuracy, the Benete team developed a taxonomy that details each JSON property’s meaning and value and a mapping ruleset. Using the instruction-tuned Llama model, they specified the desired style, tone and length of the summaries.
THEIR SUCCESS
Enabling caregivers to accomplish more for elderly clients’ health
The automated summary reports save time and improve communication across a client’s care team. They allow caregivers to focus on preventative care — an essential part of helping people live longer, healthier lives. For Benete, the reports boost the value of their solution.Projected results:
Less than 5 seconds to generate a client summary report
30 minutes saved on understanding charts per nurse per day
“Open-source Llama allowed us to run models locally, helping us ensure data privacy while also maintaining portability and control over the system.”
Kari Bäckman, CEO, Benete
“Llama proved to be the best choice for us. The Llama model outperformed our other option, a Mistral model, for accuracy and relevance while also delivering a competitive edge in average generation time and inference cost. Llama was 13.3% faster per response and produced 4.3x more tokens per dollar.”
Kari Bäckman, CEO, Benete
“The Llama-enabled summary solution not only enhances the value we deliver to customers, but it also positions us at the forefront of AI innovation in the care industry.”
Shopify | Llama case studiesShopify uses Llama to generate product pages, localize content, and automate support, helping developers scale workflows and save time.Read more
ConsumerTech
Scribd, INC | Llama case studiesDelivering faster, cheaper and more accurate results with a Llama-powered AI content discovery assistant.Read more
Tech
Exati | Llama case studiesTransforming support for smart city platform clientsRead more
A Llama-powered, automated client sleep summary provides health insights.
50% estimated increase in the quality and quantity of reported data
*All results are self-reported and not identifiably repeatable. Generally expected individual results will differ.
Stay up-to-date
Our latest updates delivered to your inbox
Subscribe to our newsletter to keep up with the latest AI updates, releases and more.
