Shares of artificial intelligence company, Cerence Inc. (NASDAQ: CRNC), doubled on Thursday based on solid Q4 earnings per share.
This positive change was a considerable deviation from what was originally an expected loss per share, which indicated better financial health than originally anticipated. Cerence also exceeded expectations in Q4 with USD$54.8 million in revenue, despite a year-over-year decline, showcasing effective management of its operations amid market challenges.
During the earnings call, the company announced a strategic pivot towards generative AI. It aims to drive future growth by leveraging this technology for automotive applications. This focus on innovation likely boosted investor confidence.
Cerence also outlined its vision for achieving profitability by 2025, centered on generative AI. This clear, forward-looking strategy appears to have significantly contributed to the stock’s increase, as investors responded positively to the promise of near-term profitability.
“Throughout the course of the year, we gained critical momentum for our generative AI- and large language model-based solutions, with six generative AI program launches with leading automakers in fiscal 2024,” said Brian Krzanich, Chief Executive Officer of Cerence.
The company originally projected FY25 revenue between USD$236 million and USD$247 million, surpassing consensus estimates. This optimistic forecast likely reassured investors about the company’s growth potential, driving the stock price higher.
Furthermore, the company launched 22 platforms. These new platforms demonstrate continued product development and market expansion, which likely boosted investor confidence in its market position and future revenue streams.
Cerence reported positive cash flow from operations at USD$6.1 million, displaying effective financial management and signalling strong liquidity and operational efficiency to shareholders.
Read more: Quantum Computing shares spike over 40% on partnership with University of Texas
Read more: Sick of telescammers pestering you? Now there’s an ‘AI Granny’ designed to waste their time
Cerence introduced an large language model for vehicles
Cerence Inc. is a multinational software company specializing in artificial intelligence technologies for the automotive industry. Founded in 2019 as a spin-off from Nuance Communications’ automotive division, Cerence develops AI-powered solutions that enhance interactions between drivers and their vehicles
Earlier this month, Cerence introduced CaLLM Edge, an automotive-grade, embedded small language model (SLM) designed to power its AI assistant platform. Integrated into Cerence’s existing solutions portfolio, CaLLM Edge ensures an intelligent, seamless user experience regardless of connectivity. Developed and optimized in collaboration with Microsoft, the model is available to Cerence’s automaker customers and listed in the Microsoft Azure AI model catalogue.
Cerence fine-tuned CaLLM Edge on Microsoft’s Phi-3 family of small language models. It uses its extensive automotive dataset to deliver specialized AI capable of handling various automotive use cases.
With 3.8 billion parameters, 4k context size, and 4-bit quantization, the model fits efficiently within an automotive head unit. Its capabilities include processing implicit and explicit car control commands, such as temperature controls and operating windows and doors.
CaLLM Edge is compatible with major automotive platforms and supports multiple deployment models. It functions in embedded-only setups, allowing operation without connectivity. Additionally, it operates in hybrid or cloud-first environments, where it acts as a backup when connectivity is unavailable.
This setup ensures users have always-on access to generative AI features and improved data privacy. Subsequently, data remains onboard rather than being sent to the cloud. For automakers, CaLLM Edge enhances assistant performance and reduces costs by leveraging an embedded SLM model.
.
