This app uses on-device machine learning to filter unwanted messages without requiring network access. It intelligently identifies spam by analyzing message semantics, eliminating the need for manual rule configuration. Users can contribute to its learning by reporting new message samples.
Summarized from real user reviews of 熊猫吃短信. Based on 9.1K reviews and a 4.5/5 average, it highlights the 5 positives and 3 negatives reviewers mention most about this Utilities app.
2.44
October 15, 2025
添加 iOS 26 的信息修改说明： 1. 信息显示逻辑的修改 信息 App 首页目前将只显示【已知联系人】的短信，其它短信都将显示在二级甚至更深的页面中。 被过滤的短信下方将不再显示"由熊猫吃短信过滤" 2. 短信通知逻辑的修改 iOS 26 改变了通知的逻辑，默认只有验证码会通知，其它短信都不会再通知，需要在「允许通知」里面选中「个人」和「交易信息」后，其它的非垃圾短信才会提醒。
More2 issues
Verification codes being incorrectly filtered
Persistent notification badges even after filtering
4 requests
iCloud backup and sync for settings
Apple Watch integration for message filtering
熊猫吃短信 is an application designed to filter unwanted spam messages from your SMS inbox. It utilizes on-device machine learning to analyze the semantic meaning of messages, allowing it to accurately identify and block spam without requiring manual rule configuration. The app operates entirely offline, prioritizing user privacy.
The app employs Core ML, a machine learning framework, to process messages directly on your device. It learns from a vast dataset of SMS samples to recognize the semantic features of spam, differentiating them from legitimate messages, public service announcements, and advertisements. This intelligent analysis means you don't need to set up any keyword filters yourself.
Yes, 熊猫吃短信 is a free application. Based on the provided information, there are no in-app purchases or subscription fees associated with its use. The app is designed to be accessible to all users looking for an effective spam filtering solution.
The app is designed with a strong emphasis on user privacy and operates completely offline, not requesting network permissions by default. It does not collect any of your personal information or upload your SMS content. If you voluntarily choose to report a spam message as a sample to help the app learn, only the anonymized text of that message and your designation of it as spam will be collected.
You can contribute to the app's intelligence by 'feeding' it new message samples directly within the application. When the app makes an incorrect judgment, reporting the message helps to expand its learning dataset. The more samples users provide, the smarter the app becomes at identifying a wider range of spam scenarios.
