Aquesta és una versió anterior. La versió actual és la 0.8.21. Ekko Studio 0.8.21 · Data de publicació:
Fitxers d’instal·lació 0.7.31
| Arquitectura | Tipus | Àmbit | Mida del fitxer | Baixa | SHA256 |
|---|---|---|---|---|---|
| 64 bits | EXE (NSIS) | Usuari actual | — | github.com | VirusTotal |
| 64 bits | EXE (NSIS) | Tots els usuaris | — | github.com | VirusTotal |
L’enllaç i l’SHA256 provenen del manifest comunitari del Gestor de paquets del Windows. Convé comparar el hash del fitxer baixat abans d’executar-lo.
Notes de la versió
更新重点
- 新增 App 与 Studio 的鉴权 P2P 直连,直连不可用时自动回退云端中转 (#3290)
- 改善多网卡和 Docker 环境的 P2P 连接,并修复代理 fake-IP DNS 导致的直连失败 (#3292、#3303)
- 统一模型上下文、推理强度和计费信息识别,改进自定义接口及带前缀模型 ID 的匹配,并优先使用手动设置的价格 (#3298、#3300)
- 修复 App 自定义接口的上下文查询,并为共享会话补充模型别名与推理选项 (#3299)
- 修复刷新或重启后丢失漫画主题选择的问题,加载时恢复已保存的主题风格 (#3302)
- Studio 与桌面端版本同步更新为 0.7.31,前端更新日志补齐 11 种语言并保留最近三个版本 (#3305)
What's Changed
- [codex] add authenticated direct App transport with relay fallback by @EKKOLearnAI in https://github.com/EKKOLearnAI/ekko-studio/pull/3290
- [codex] bind P2P sockets to uplinks and publish Docker UDP ports by @EKKOLearnAI in https://github.com/EKKOLearnAI/ekko-studio/pull/3292
- [codex] unify model metadata and pricing resolution by @EKKOLearnAI in https://github.com/EKKOLearnAI/ekko-studio/pull/3298
- [codex] forward App context endpoints and shared reasoning metadata by @EKKOLearnAI in https://github.com/EKKOLearnAI/ekko-studio/pull/3299
- [codex] match custom model IDs by final path segment by @EKKOLearnAI in https://github.com/EKKOLearnAI/ekko-studio/pull/3300
- [codex] preserve selected theme style on reload by @EKKOLearnAI in https://github.com/EKKOLearnAI/ekko-studio/pull/3302
- [codex] fix P2P STUN resolution under fake-IP DNS by @EKKOLearnAI in https://github.com/EKKOLearnAI/ekko-studio/pull/3303
- [codex] prepare Studio 0.7.31 with localized changelog by @EKKOLearnAI in https://github.com/EKKOLearnAI/ekko-studio/pull/3305
Full Changelog: https://github.com/EKKOLearnAI/ekko-studio/compare/v0.7.30...v0.7.31https://github.com/EKKOLearnAI/ekko-studio/releases/tag/v0.7.31