[GH-ISSUE #1620] 能否可以将文档整理一下，第一次学习真的有点费劲 #1053

New issue

Closed

opened 2026-10-29 14:48:00 +00:00 by bsakaguchi · 3 comments

bsakaguchi commented

2026-10-29 14:48:00 +00:00

Owner

Originally created by @Neverland53 on GitHub (Nov 15, 2025).
Original GitHub issue: https://github.com/kvcache-ai/ktransformers/issues/1620

Originally assigned to: @KMSorSMS on GitHub.

Reminder

I have read the above rules and searched the existing issues.

Description

是否可以将文档整理一下，第一次学习真的有点费劲。
how to start、installation =有点乱。

Pull Request

No response

Originally created by @Neverland53 on GitHub (Nov 15, 2025). Original GitHub issue: https://github.com/kvcache-ai/ktransformers/issues/1620 Originally assigned to: @KMSorSMS on GitHub. ### Reminder - [x] I have read the above rules and searched the existing issues. ### Description 是否可以将文档整理一下，第一次学习真的有点费劲。 how to start、installation =有点乱。 ### Pull Request _No response_

bsakaguchi

2026-10-29 14:48:00 +00:00

closed this issue
added the
enhancement
label

bsakaguchi commented

2026-10-29 14:48:00 +00:00

Author

Owner

@KMSorSMS commented on GitHub (Nov 16, 2025):

Have you checked the doc: https://kvcache-ai.github.io/ktransformers/en/kt-kernel_intro.html? If you want to use the SFT part, I understand there is some chaos. You can check the pr#1622 for the docs structure change. As the kt-kernel part is still ongoing with heavy work, the tutorial may come when I am available to write a tech guide for developers. (Sorry for the delay :( )

@KMSorSMS commented on GitHub (Nov 16, 2025): Have you checked the doc: https://kvcache-ai.github.io/ktransformers/en/kt-kernel_intro.html? If you want to use the SFT part, I understand there is some chaos. You can check the pr#1622 for the docs structure change. As the kt-kernel part is still ongoing with heavy work, the tutorial may come when I am available to write a tech guide for developers. (Sorry for the delay :( )

bsakaguchi commented

2026-10-29 14:48:00 +00:00

Author

Owner

@Neverland53 commented on GitHub (Nov 16, 2025):

请问大佬，如果是qwen3-30B之外的模型，也是按照这kt-kernel readme中的方式进行部署嘛

我看还有很多文档部署deepseek qwen3-235 之类的，是否都作废，按照 kt-kernel readme 中的方式进行部署呢。
此外我看咱们release已经发了0.4.1，这个有相关的使用教程嘛。

@Neverland53 commented on GitHub (Nov 16, 2025): 请问大佬，如果是qwen3-30B之外的模型，也是按照这kt-kernel readme中的方式进行部署嘛我看还有很多文档部署deepseek qwen3-235 之类的，是否都作废，按照 kt-kernel readme 中的方式进行部署呢。此外我看咱们release已经发了0.4.1，这个有相关的使用教程嘛。

bsakaguchi commented

2026-10-29 14:48:00 +00:00

Author

Owner

@KMSorSMS commented on GitHub (Nov 16, 2025):

对的，以前的一些部署推理文档是用在 archive 的那个文件夹的那一套方式，现在为了更好的兼容性和对更多模型的支持，就参考 kt-kernel readme 的方式部署即可。0.4.1 的包是属于微调的，目前 kt 推理部分的操作建议直接 main 分支源码编译，不过后续会加上 docker 镜像方便安装

@KMSorSMS commented on GitHub (Nov 16, 2025): 对的，以前的一些部署推理文档是用在 archive 的那个文件夹的那一套方式，现在为了更好的兼容性和对更多模型的支持，就参考 kt-kernel readme 的方式部署即可。0.4.1 的包是属于微调的，目前 kt 推理部分的操作建议直接 main 分支源码编译，不过后续会加上 docker 镜像方便安装