Improvements
- When a model is rejected for image input, the app now explains why it is treated as text-only and points to the Accepts images setting.
- On Linux, the window title and desktop entry now have matching names, and the app icon is installed.
- Environment variables from the login shell are now only used when exported, and Codex proxy details show which variable was read.
Bug Fixes
- Fixed an issue where assistant replies cut off before generating any content could still be sent to upstream chat services.
Install
Download from usemagpie.ai, or curl -fsSL https://usemagpie.ai/install.sh | sh. An installed magpie updates itself.
- macOS (Apple Silicon):
magpie-darwin-arm64.dmg - macOS (Intel):
magpie-darwin-amd64.dmg - Windows:
magpie-windows-amd64.exe(ARM:magpie-windows-arm64.exe) - Linux:
magpie-linux-amd64/magpie-linux-arm64(needs GTK 3 and WebKitGTK 4.1) - Terminal only:
magpie-cli-<os>-<arch>
改进
- 当模型因图像输入被拒绝时，应用现在会解释为什么将其视为仅文本，并指向 接受图像 设置。
- 在 Linux 上，窗口标题和桌面条目现在具有匹配的名称，并且应用图标已安装。
- 来自登录 shell 的环境变量现在仅在导出时使用，并且 Codex 代理详细信息会显示读取了哪个变量。
错误修复
- 修复了助手回复在生成任何内容之前就被截断但仍可能被发送到上游聊天服务的问题。
安装
从 usemagpie.ai 下载，或运行 curl -fsSL https://usemagpie.ai/install.sh | sh。已安装的 magpie 会自动更新。