Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Cline插件调用卡在Api request #265

Open
DZCD opened this issue Jan 13, 2025 · 3 comments
Open

Cline插件调用卡在Api request #265

DZCD opened this issue Jan 13, 2025 · 3 comments

Comments

@DZCD
Copy link

DZCD commented Jan 13, 2025

经过排查,发现当大量输入token时,回复等待时间超过10分钟!

@Mattchew86
Copy link

这整天对我来说都是这样。周五晚上效果非常好,我完成了很多事情。但今天完全无法使用。我本以为可能是他们的API问题,但我有一个单独的Python脚本使用他们的API和不同的密钥在运行,没有任何问题。我尝试为cline添加一个新的API密钥,仍然没有用。我还试了通过open router使用deepseek v3,它可以工作,但速度极其缓慢。可能是cline的IP被屏蔽或限速了?

cline/cline#1157

@mowentian
Copy link
Contributor

亲爱的 DeepSeek V3 用户和 Cline 社区成员,

感谢大家对 DeepSeek V3 的支持和反馈。我们注意到最近在使用 Cline + DeepSeek API 时,部分用户遇到了延时较高的问题。其成因主要是长上下文窗口及多并发引起的队列,为了优化使用体验,我们整理了一些建议:

  1. 关闭 Cline 的 Auto Approve 功能:建议在使用 DeepSeek 时,关闭 Cline 的 Auto Approve 功能,以降低资源消耗,降低多请求因并发带来的卡顿风险。

  2. 使用 Cline 改版:例如 Roo-Cline 或 ALine。这些改版携带的上下文相对短,处理请求时表现更为流畅,且缓存命中率更高,可以帮助节省 API 调用成本。

  3. 慎用 DeepSeek 处理大型工程:对于大型代码工程,不建议使用 DeepSeek + Cline。大型工程通常会携带大量代码上下文,会导致 API 请求处理速度明显变慢。

  4. 更换其他 LLM 提供商:建议您在使用 Cline 时,更换其他 LLM 提供商,以获得更流畅的体验。

我们深知这些问题的存在给大家带来了不便,团队正在积极优化调度算法和服务器负载管理,以提升整体性能和稳定性。感谢大家的理解与支持,我们将继续努力为大家提供更好的服务。

Dear Cline Users,
Thank you for your continued support and feedback on DeepSeek V3. We have noticed that some users have recently experienced higher latency when using Cline + DeepSeek API. The primary causes are long context windows and multi-concurrency queues. To optimize your experience, we have compiled some recommendations:

  1. Disable Cline's Auto Approve Feature: It is recommended to turn off Cline's Auto Approve feature when using DeepSeek to reduce resource consumption and minimize the risk of lag caused by multiple concurrent requests.

  2. Use Modified Versions of Cline: For example, Roo-Cline or ALine. These versions have shorter context lengths, handle requests more smoothly, and have higher cache hit rates, which can help save API call costs.

  3. Avoid Using DeepSeek for Large Projects: For large code projects, it is not recommended to use DeepSeek + Cline. Large projects typically involve extensive code contexts, which can significantly slow down API request processing.

  4. Switch to Other LLM Providers: We recommend trying other LLM providers when using Cline for a smoother experience.

We understand that these issues have caused inconvenience, and our team is actively working on optimizing scheduling algorithms and server load management to improve overall performance and stability. Thank you for your understanding and support. We will continue to strive to provide you with better service.

Best regards,
The DeepSeek Team

@anoyi09
Copy link

anoyi09 commented Jan 15, 2025

就你们家便宜,其他的API用不起,希望能解决卡顿问题

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants