mirror of
https://github.com/kvcache-ai/ktransformers.git
synced 2026-11-05 01:07:28 +00:00
[GH-ISSUE #602] Windows,Turing架构的显卡部署有问题 #442
Labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
kvcache-ai/ktransformers#442
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @zhizi42 on GitHub (Feb 22, 2025).
Original GitHub issue: https://github.com/kvcache-ai/ktransformers/issues/602
使用预编译好的whl包能安装,但是使用命令启动时报错,翻阅issue发现疑似本来不支持我的Turing的2070s,看到解决方案为改代码自己编译,遂尝试自己编译,但遇到以下报错,翻阅Google和本项目issue无法战胜。不知道为什么会中文乱码,试了chcp 65001和改各种编码都没用。
@LemonFan-maker commented on GitHub (Feb 22, 2025):
试过WSL2了没有?别用windows环境跑,用wsl2的22.04跑一下试试?
@ubergarm commented on GitHub (Feb 22, 2025):
@zhizi42
RTX 2070 Super 仅配备 8GB 显存,而 R1 模型最低需求约为 14GB(建议至少 16GB)。对于较小模型或手动配置
--optimize_config_path ktransformers/optimize/optimize_rules/DeepSeek-V3-Chat.yaml的情况,暂不确定是否可行(但这一步骤需在成功编译安装后才有意义)。
我完全同意 @LemonFan-maker 的建议:强烈推荐尝试 WSL2 环境,并参考
此构建指南。该指南不仅提供最新的预编译
.whl文件,还附有完整操作说明。
@zhizi42
RTX 2070 Superonly has 8GB VRAM and at least for R1 the minimum is around 14GB though I suggest at least 16GB. I am not sure for smaller models or if you manually configure the--optimize_config_path ktransformers/optimize/optimize_rules/DeepSeek-V3-Chat.yamlhowever this only matters after you get it to compile and install.@c00jsw00 commented on GitHub (Feb 25, 2025):
請使用WSL 在win上安裝LINUX (在安裝ktransformers)
https://bayareanotes.com/wsl-installation/
@knilink commented on GitHub (Mar 3, 2025):
chrono头文件问题,应该已经修复https://github.com/kvcache-ai/ktransformers/pull/622
@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!