[GH-ISSUE #231] Is it possible to reduce the CPU memory requirements by modifying the configuration？ #137

New issue

Closed

opened 2026-10-29 14:34:50 +00:00 by bsakaguchi · 4 comments

bsakaguchi commented

2026-10-29 14:34:50 +00:00

Owner

Originally created by @ChenyangRan on GitHub (Feb 13, 2025).
Original GitHub issue: https://github.com/kvcache-ai/ktransformers/issues/231

Thank you for the tutorials on deploying DeepSeek R1. I would like to deploy the DeepSeek R1 Q4 version on my machine, but the problem is that I only have 256G RAM, the good news is that my GPU memory is sufficient or even more. Is it possible to move some of the CPU calculations to the GPU, and if so what should I do?

Originally created by @ChenyangRan on GitHub (Feb 13, 2025). Original GitHub issue: https://github.com/kvcache-ai/ktransformers/issues/231 Thank you for the tutorials on deploying DeepSeek R1. I would like to deploy the DeepSeek R1 Q4 version on my machine, but the problem is that I only have 256G RAM, the good news is that my GPU memory is sufficient or even more. Is it possible to move some of the CPU calculations to the GPU, and if so what should I do?

bsakaguchi

2026-10-29 14:34:50 +00:00

closed this issue
added the
stale
label

bsakaguchi commented

2026-10-29 14:34:51 +00:00

Author

Owner

@Radohead commented on GitHub (Feb 15, 2025):

Please read faq

@Radohead commented on GitHub (Feb 15, 2025): Please read faq

bsakaguchi commented

2026-10-29 14:34:51 +00:00

Author

Owner

@vc5409ftu commented on GitHub (Feb 16, 2025):

256内存就需要32g 显存

@vc5409ftu commented on GitHub (Feb 16, 2025): 256内存就需要32g 显存

bsakaguchi commented

2026-10-29 14:34:51 +00:00

Author

Owner

@ChenyangRan commented on GitHub (Feb 17, 2025):

256内存就需要32g 显存

方便将对应的yaml分享出来作为参考吗？

@ChenyangRan commented on GitHub (Feb 17, 2025): > 256内存就需要32g 显存方便将对应的yaml分享出来作为参考吗？

bsakaguchi commented

2026-10-29 14:34:51 +00:00

Author

Owner

@ErvinXie commented on GitHub (Dec 11, 2025):

This issue has been automatically closed due to inactivity for 60 days.

If you believe this issue is still relevant, please feel free to reopen it with additional information or context.

Thank you for your contribution to KTransformers!

此 issue 因超过 60 天无活动而被自动关闭。

如果您认为此问题仍然存在，请随时重新打开并提供更多信息或上下文。

感谢您对 KTransformers 的贡献！

@ErvinXie commented on GitHub (Dec 11, 2025): This issue has been automatically closed due to inactivity for 60 days. If you believe this issue is still relevant, please feel free to reopen it with additional information or context. Thank you for your contribution to KTransformers! --- 此 issue 因超过 60 天无活动而被自动关闭。如果您认为此问题仍然存在，请随时重新打开并提供更多信息或上下文。感谢您对 KTransformers 的贡献！