Skip to content

Issues: kvcache-ai/ktransformers

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.

Sign up for GitHub

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

401 Open 208 Closed
401 Open 208 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

[Bug] the chat responses are repeated
#846 opened Mar 9, 2025 by learner-crapy
5 tasks done
1
[Feature] 提供官方docker hub 镜像和docker-compose.yml文件
#843 opened Mar 7, 2025 by verigle
2 tasks done
2
目前用4张4090 48G的+256内存跑1.58bit量化版本是不是最省钱的多并发方案？
#841 opened Mar 7, 2025 by lifeha
编译成功Ktransformers后，无法运行，提示：cannot find -lcuda: 没有那个文件或目录
#838 opened Mar 7, 2025 by KimmyGLM
[Bug] v0.2.3 FP8 线性内核 4090报错f8e4m3nv is only supported on compute capability >= 90
#836 opened Mar 7, 2025 by pitaozi
5 tasks
@Azure-Tang
1
2T 内存 + 单张L40 显卡，能跑起满血版非量化的deepseek-R1 不？
#832 opened Mar 7, 2025 by hlj1205
17
[Bug] make dev_install编译失败
#831 opened Mar 7, 2025 by wangfan008
5 tasks
1
[Bug] No Tensor Saved when Using merge_safetensor_gguf.py
#830 opened Mar 7, 2025 by 3wweiweiwu
5 tasks done
@Azure-Tang
1
[Bug] Cascade Lake架构不支持AVX512VPOPCNTDQ，0.2.3编译失败
#829 opened Mar 6, 2025 by chenmz00
5 tasks done
@KMSorSMS
6
【性能报告】Windows Server下双路9654+双路3090性能汇报
#828 opened Mar 6, 2025 by PC-DOS
7
[Feature] QwQ-32B支持
#827 opened Mar 6, 2025 by HuntZhaozq
2 tasks
7
运行的显存有计算方式吗？
#826 opened Mar 6, 2025 by tzhang2014
2
[Bug] 使用DeepSeek-R1-UD-IQ1_M进行聊天，回答输出为空
#825 opened Mar 6, 2025 by freepanzer
3 of 5 tasks
1
Ktransformer 跑DeepSeek-R1-Q4_K_M ,Intel(R) Xeon(R) Gold 6330 CPU ，双槽（56核），推理速度 仅 7个tokens/s 。还有提升空间不？
#824 opened Mar 6, 2025 by hlj1205
9
大佬们问个简单的问题，源码可以直接运行，install安装的意义是什么
#823 opened Mar 6, 2025 by WingsLong
1
【性能分析】这几天研究了一下nsys profile，尝试分析目前的generate时间瓶颈，分享下结论，不保证正确，希望各位讨论
#821 opened Mar 6, 2025 by tarjintor
7
[Bug] 输出的内容中，content，其他的选项都是错误的或者null
#818 opened Mar 6, 2025 by huqianghui
5 tasks done
1
[Bug] 无法将ktransformers集成到dify中
#817 opened Mar 6, 2025 by magic-hya
5 tasks done
1
RuntimeError: CUDA error: unknown error; Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.
#816 opened Mar 6, 2025 by meda0719
@Atream
2
[Feature] 针对多路系统实现张量并行
#812 opened Mar 5, 2025 by yeungtuzi
2 tasks
7
[Feature] 支持AMD GPU
#807 opened Mar 5, 2025 by yeungtuzi
2 tasks
@Azure-Tang
1
单卡RTX 4090 + CPU部署deepseek-r1:671b(int4),测试token值为4.1
#806 opened Mar 5, 2025 by steven-RD
6
[Bug] 缓存未释放导致达到一定推理次数后推理速度降低
#805 opened Mar 5, 2025 by Pb-207
5 tasks done
@Atream
[Bug] RuntimeError: CUDA error: invalid device function
#801 opened Mar 5, 2025 by kaixwangwei
1 of 5 tasks
@Atream
3
[Report] EPYC 9135*2 , 24*48G DDR5 6400, 4090, SUCCESSFUL
#797 opened Mar 5, 2025 by opernal
3
ProTip! Updated in the last three days: updated:>2025-03-06.