Cline 4.1.25 新增云端会话,逐步推出

Cline Cloud 会话逐步推出,可在托管沙盒中对你选择的 GitHub 仓库和分支运行任务。

Cline

事件类型: 模型与平台
版本: 4.1.25
官方发布: 2026-10-16
来源: github.com/cline/cline/releases/tag/v4.1.25
本站整理: 2026-10-17

本次变化

共 5 条要点

Cline Cloud 会话正在逐步推出,可在托管沙盒中对你选择的 GitHub 仓库和分支运行任务。云端任务会在 Recent 和 History 中显示状态,完成时通知你,关闭编辑器后仍继续运行。
官方原文 · c1 c2
Added Cline Cloud sessions, rolling out gradually: run a task in a hosted sandbox on a GitHub repo and branch you pick.
官方原文 · c1
Cloud tasks show in Recent and History with their status, notify you when they finish, and keep running after you close your editor.
官方原文 · c2
未选择模型时,Cline 提供方现在使用 Gemini 4 Argon,它是 Cline 推荐列表中新的第一个模型。OpenRouter 仍使用 Claude Sonnet 5.5。Cline 还新增了 Boundless 和 Standard Compute 两个提供方。
官方原文 · c3 c4 c9
With no model chosen, the Cline provider now uses Gemini 4 Argon, the new first model in Cline's recommended list.
官方原文 · c3
OpenRouter stays on Claude Sonnet 5.5.
官方原文 · c4
Adds the Boundless and Standard Compute providers
官方原文 · c9
发送给本地模型(LM Studio、Ollama)的长请求,在服务器仍在处理大型提示词时,不再因 5 分钟无响应而以 terminated: BodyTimeoutError 失败。
官方原文 · c5
Long requests to local models (LM Studio, Ollama) no longer fail after 5 minutes of silence with terminated: BodyTimeoutError while the server is still processing a large prompt.
官方原文 · c5
OpenAI Compatible 的基础 URL 现在按输入原样使用。提供方错误在响应中途打断时,Cline 最多重试 3 次,之后才使该轮失败。
官方原文 · c6 c7
OpenAI Compatible base URLs are used exactly as entered.
官方原文 · c6
When a provider error interrupts a response mid-stream, Cline retries it up to 3 times before failing the turn.
官方原文 · c7
附在 Cline 问题回答中的图片和文件现在会送达模型,不再被丢弃。
官方原文 · c8
Images and files attached to an answer to one of Cline's questions now reach the model instead of being dropped.
官方原文 · c8

变化明细

下表字段直接取自官方原文，保持原文语言，不做翻译。

编号	主体	动作	对象	范围	数值	判断
c1	Cline	added	Cline Cloud sessions	rolling out gradually; hosted sandbox on a GitHub repo and branch you pick
c2	Cloud tasks	show, notify, keep running	status in Recent and History, finish notification, running after editor closes
c3	Cline provider	uses by default	Gemini 4 Argon	when no model is chosen
c4	OpenRouter	stays on	Claude Sonnet 5.5
c5	Long requests to local models (LM Studio, Ollama)	no longer fail	after 5 minutes of silence with terminated: BodyTimeoutError	while the server is still processing a large prompt	精确 5 minutes	否定
c6	OpenAI Compatible base URLs	are used	exactly as entered
c7	Cline	retries	a response interrupted mid-stream by a provider error	before failing the turn	最多 3 times
c8	Images and files attached to an answer to one of Cline's questions	now reach	the model instead of being dropped
c9	Cline	adds	the Boundless and Standard Compute providers