Bug Fixes
- Fixed an issue where images could be sent directly to text-only models when an image describer was enabled. Models not known to support images now receive a text description of the image instead.
- Fixed image description requests not appearing clearly in conversation routing. The description now shows as its own Vision step tied to the relevant model and session.
Improvements
- Image descriptions are now handled consistently across all supported integrations, including Responses, Codex, Chat, Anthropic, and Gemini.
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>
Bug Fixes
- 修复了启用图像描述器时，图像可能被直接发送给纯文本模型的问题。未明确支持图像的模型现在会收到该图像的文本描述。
- 修复了图像描述请求在对话路由中无法清晰显示的问题。该描述现在会作为独立的 Vision 步骤显示，并与相关模型和会话绑定。
Improvements
- 现在，图像描述在所有受支持的集成中统一处理，包括 Responses、Codex、Chat、Anthropic 和 Gemini。
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。