Skip to content

v2.1.24-beta.b2c2835

Pre-release
Pre-release

Choose a tag to compare

@github-actions github-actions released this 09 Nov 12:39
· 11 commits to dev since this release

✨ 本次更新

  • 新增「安全策略熔断」功能,自动拦截已被上游拒绝的违规对话,防止因重复提交违规内容导致账号被封。
  • 优化桌面端更新体验,支持后台静默下载并在控制台顶栏实时显示进度,支持一键重启完成安装。
  • 设置页新增「允许局域网访问」开关,无需手动修改配置即可让同网络下的其他设备连接。
  • 增强弱网环境下的连接稳定性,支持自动重试传输失败的请求并优化了超时处理机制。
  • 首页新增 Antigravity 账号管理面板,支持 Google OAuth 快捷登录及详细的额度用量展示。
  • 支持自动从云端同步最新的模型计费价格表,并新增对 gemini-2.5-pro 等模型的支持。
  • 补齐图片编辑(/v1/images/edits)接口支持,兼容 OpenAI 标准的文件上传与 JSON 协议。

What's New

  • Added "Safety Policy Circuit Breaker" to automatically block conversations already rejected by upstream filters, preventing account bans from repeated violations.
  • Enhanced the desktop update experience with background downloads and real-time progress tracking in the dashboard for one-click installation.
  • Added an "Allow LAN Access" toggle in settings, making it easy to connect other devices on the same network without manual configuration.
  • Improved connection stability in weak network conditions with automatic retries for failed transmissions and optimized timeout handling.
  • Introduced an Antigravity account management panel on the home page, supporting Google OAuth login and detailed usage tracking.
  • Enabled automatic synchronization of the latest model pricing tables and added support for new models including gemini-2.5-pro.
  • Added support for the image editing (/v1/images/edits) endpoint, compatible with standard OpenAI file upload and JSON formats.
完整提交记录 / Full commit list
  • feat(policy-guard): stop resending rejected conversations upstream (#876)