88Z AI, 29 April — DeepSeek's multimodal 'Image Recognition Mode' has entered beta testing, with a new entry point visible on the homepage for some users. This feature offers human-like visual understanding, accurately identifying objects, scenes, and visual illusions, while also pinpointing specific locations like Hangzhou's Lingyin Temple through architectural and textual clues.
At a glance
- DeepSeek's image recognition mode enters beta, with new entry visible to some users
- Features human-like visual understanding to accurately identify species, locations, and visual illusions
The update
DeepSeek has officially launched the beta testing of its multimodal 'Image Recognition Mode,' with some users now seeing a new feature entry on the app's homepage. In this mode, the system simulates human visual cognition to deeply understand object details and overall scenes in uploaded images. Practical tests demonstrate high recognition accuracy. The mode can correctly identify animal species and effectively detect various visual illusion images. Furthermore, by analyzing architectural styles and text within images, the system successfully identified Hangzhou's Lingyin Temple and provided precise geographic coordinates, showcasing its comprehensive analytical capabilities in complex scenarios.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- 智东西Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.