OPPO's First AI Large Model! AndesGPT Announced
-
According to OPPO's official announcement, the company held a technical conference for the Find X7 series today. At this event, OPPO's first AI large model, AndesGPT, was officially released.
According to official sources, AndesGPT features a comprehensive and powerful generative vision model that can accurately understand the semantics of images, and on this basis, it supports the recognition and segmentation of over 120 categories of subjects.
Not only does it achieve high-precision segmentation down to hair-level detail, but it can also support the segmentation and extraction of up to six subjects. Additionally, it can fill and generate ultra-large images, even perfectly recreating natural lighting and shadows in pictures.
Image generation time locally is as low as 6 seconds, leading other models on the same platform by 60%, allowing edge devices to achieve generation precision and quality comparable to cloud-based solutions.
Additionally, compared to a 1-billion-parameter large model, the AndesGPT 7-billion-parameter language model offers more "intelligent" comprehension capabilities.
For example, in the call summarization feature, AndesGPT can help users organize the key points of call content and accurately output summaries with themes, key points, and to-do items, significantly outperforming 1-billion-parameter edge-side models in accuracy.
