Changelog:
- Changed styling to a navigation bar format.
- Fixed an issue where pressing C and Shift while using the magnifying glass had no effect.
- special thanks to @thiagodalsoto for providing Brazilian Portuguese translations
- Additional bug fixes and experience improvements.
Q&A:
Linux / Mac
-
Q1: Linux won't run after installing a new version?
Uninstall Sunny version 1.x or earlier, and try removing old configuration filesrm -rf ~/.config/Sunny/*
-
Q2: On Ubuntu 24.04, screenshots capture a white desktop. Why?
To support more Linux distributions, Sunny 1-2.x currently supports onlyx11
and does not supportWayland
(Ubuntu's default). Switch tox11
with the following steps (details in #55:sudo vim /etc/gdm3/custom.conf # Uncomment the line: WaylandEnable=false, then save the file. sudo systemctl restart gdm3 # Reboot the system
-
Q3: How can I use Sunny on non-Debian-based distributions?
Please download and use thesunny_x64.AppImage
package. For a list of supported Linux distributions, refer to supportp_platform. -
Q4: How can I download Sunny for ARM and LOONGSON chips?
You can download v1.6.2. When build machines become available, version 2.x will also be uploaded. -
Q5: Why can’t Sunny run on my Mac?
Currently, Sunny supports only Intel-based Macs and does not support M-series chips. -
ps: The next version is expected to support Mac's M-series chips and the Linux Wayland architecture.
Offline OCR (CPU & GPU)
Extract text from images using a local offline engine.
- No registration or key input required.
- Fully offline, with all processing done locally.
- Supports both CPU and GPU modes:
- CPU Version: Highly compatible, uses less memory, processes single images quickly but is slower for batch processing (recommended for general users).
- GPU Version: Supports NVIDIA GPUs, uses more memory, but processes batch images significantly faster (1/2 to 1/3 the time of the CPU version, recommended for advanced users with NVIDIA cards).
- Only supports 64-bit Windows systems. Both versions handle single and batch image recognition. Simply drag and drop images into the window.
日志变更:
- 设置样式改为导航栏样式
- 修复使用放大镜时,按下 C 和 Shift 不生效
- 感谢 @thiagodalsoto 提供葡萄牙语-巴西语翻译
- 更多bug修复和体验优化...
常见答疑
Linux / Mac
-
Q1:Linux 安装新版后,无法运行?
卸载低 1.x 版本,且尝试运行删除旧的配置文件
rm -rf ~/.config/Sunny/*
-
Q2:在 Ubuntu 24.04 截屏,捕捉的桌面为白色?
Sunny 1-2.x 版本为兼容更多的 Linux 发行版,当前仅支持
x11
,不支持Wayland
(Ubuntu 默认); 切换x11
方式如下,详细参考 #55sudo vim /etc/gdm3/custom.conf # at the beginning of the line WaylandEnable=false to uncomment it, then save the file. sudo systemctl restart gdm3 // reboot the system
-
Q3: 非 Debian 系,如何使用?
请下载和使用
sunny_x64.AppImage
安装包;已知道支持的 Linux 发行版可参考 已兼容平台 -
Q4:ARM 和 LOONGSON 芯片如何下载?
可先下载 v1.6.2;若编译机器有空时,2.x版本便会构建上传
-
Q5:Mac 无法运行?
同理,对于 Mac 当前仅支持
intel
芯片,不支持M
系列芯片 -
ps: 预计下一个版本将会支持 Mac 的 M 芯片和 Linux Wayland 架构
离线 OCR (CPU & GPU)
通过本地离线引擎,离线提取图片中的文字
- 无需要任何注册和 key 的输入
- 全程断网离线本地运行和识别
- 支持 CPU 和 GPU 模式(叉腰):
- CPU 版本:通用性强,占用内存更少,对于单张图片解析快,批量图片耗时大(普通用户推荐)
- GPU 版本:仅支持 N 卡,占用内用多,但对于批量解析大量图片,耗时是 CPU 版本是 1/2 ~ 1/3 时间,很快(高级 N 卡推荐)
- 仅支持 win 64 bit 系统;均支持单张图片和批量图片识别解析,直接拖曳到窗口即可。