QwenLM/qwen.cpp

Issues

• 使用qwen.cpp对原模型进行转化为什么文件反而增大了？

  #79 opened a month ago by zzzcccxx
  0

• 'QWenConfig' object has no attribute 'intermediate_size'

  #34 opened a year ago by dingli06
  2

• qwen2 support

  #82 opened 5 months ago by bil-ash
  2

• 如何在x86架构上进行交叉编译到ARM64架构？

  #84 opened 3 months ago by yux-lab
  0

• [BUG] Qwen-1.8-Chat，用llama.cpp量化为f16，然后推理回答错乱，请问1.8在llama.cpp还不支持吗？

  #69 opened a year ago by Lyzin
  5

• 对qwen1.5模型进行lora微调，怎么加载微调的参数，怎么将微调后的模型进行转换

  #83 opened 3 months ago by Gooddz1
  0

• [BUG] 多轮对话的 prompt 应该如何构建？

  #81 opened 6 months ago by 791136190
  0

• pip install -U qwen-cpp 报错

  #58 opened a year ago by micronetboy
  3

• qwen1.5 support?

  #80 opened 9 months ago by anan1213095357
  2

• 如何将gradio架构构建的前端和qwen-cpp推理代码连接？

  #66 opened a year ago by tougeqaq
  2

• Does it support Qwen1.5 Model?

  #78 opened 10 months ago by kicGit
  8

• Python Binding之后，如何只使用cpu进行推理呢？

  #77 opened 9 months ago by zzzcccxx
  0

• Support `--gpu-layers`

  #45 opened a year ago by lindeer
  7

• Python Binding 报错

  #33 opened a year ago by xinbingzhe
  3

• python binding无法正常安装

  #43 opened a year ago by passionate11
  2

• 在MacOS，用python调用qwen_cpp载入模型进行推理，只能启动CPU，无法使用GPU。

  #54 opened a year ago by bigbigtooth
  1

• python-bind报错 ERROR: Could not build wheels for qwen-cpp, which is required to install pyproject.toml-based projects

  #52 opened a year ago by zhangzai666
  2

• qwen_cpp可以提供api接口实现web服务么

  #53 opened a year ago by zhangzai666
  1

• Qwen-7B-Chat WSL GPU Error: ankerl::unordered_dense::map::at(): key not found

  #29 opened a year ago by dlutsniper
  2

• why missing "assistant" here

  #76 opened 10 months ago by feixyz10
  0

• crash if compliing in debug mode, everything is ok if in release mode

  #75 opened 10 months ago by feixyz10
  0

• 如何下载tiktoken_cpp

  #74 opened 10 months ago by eswulei
  0

• 添加tokens生成速度

  #73 opened 10 months ago by OliverQueen1466
  0

• 请问用qwen.cpp量化后的模型如何使用optimum-benchmark进行性能基准测试,现在参照readme中所述只得到一个build文件夹，不清楚如何进行下一步的测试

  #72 opened 10 months ago by suyu-zhang
  0

• Why does `TextStreamer` hold on punctuation?

  #71 opened a year ago by Wovchena
  0

• windows 下使用qwen.cpp 问题

  #70 opened a year ago by kingpingyue
  0

• 希望团队能继续支持qwen.cpp

  #60 opened a year ago by awtestergit
  3

• 多轮会话

  #67 opened a year ago by litongjava
  0

• 💡 [REQUEST] - CPU 的 qwen-cpp 如何封装为一个 http 服务？

  #65 opened a year ago by micronetboy
  4

• 💡 [Question] - QwenCPP Python Binding 如何 支持 BLAS CPU 加速

  #64 opened a year ago by micronetboy
  2

• 💡 [Question] - <title>qwen-cpp 只使用 cpu 和 启用 cpu BLAS 加速, 在都不使用GPU的情况下，速度有多大差别？我测试没有差别

  #63 opened a year ago by micronetboy
  0

• 💡 [Question] - 您好，请教个问题，qwen-cpp BaseStreamer 如何通过std::string 构造一个　BaseStreamer？Ｃ＋＋代码少一个构造方式

  #62 opened a year ago by micronetboy
  0

• 您好，请教个问题，qwen-cpp BaseStreamer 如何通过std::string 构造一个　BaseStreamer？Ｃ＋＋代码少一个构造方式

  #61 opened a year ago by micronetboy
  0

• 为啥qwen.cpp在A100和A10性能差距很大

  #56 opened a year ago by zhangzai666
  1

• Python Binding 如何 支持BLAS CPU 加速

  #59 opened a year ago by micronetboy
  0

• Python Binding在windows下无法编译

  #37 opened a year ago by AppleJunJiang
  1

• CUDA error 2 at /home/qwen.cpp/third_party/ggml/src/ggml-cuda.cu:7196: out of memory

  #55 opened a year ago by youngallien
  0

• 72B模型量化需要多大内存，192G的内存都会被kill掉

  #47 opened a year ago by sweetcard
  9

• 请问7b的模型量化需要多大的内存，我这一直显示out of memory

  #51 opened a year ago by WCSY-YG
  0

• qwen.cpp合并到llama.cpp中之后，对于<|im_start|>、<|im_end|>似乎没有正确处理

  #50 opened a year ago by listeng
  0

• 代码ctx_w_size

  #49 opened a year ago by EveningLin
  0

• Support for AMD‘s ROCm

  #46 opened a year ago by riverzhou
  5

• 很容易出现 UnicodeDecodeError: 'utf-8' codec can't decode bytes

  #36 opened a year ago by zhcharles
  6

• GGML_ASSERT when using a long prompt

  #44 opened a year ago by Ayahuasec
  2

• Qwen-7B-Q4_0 works well on Mac M1, but Qwen-7B-Q8_0 cannot work with a ggml-metal error.

  #42 opened a year ago by songkq
  1

• Does the Owen.cpp support macOS metal build?

  #38 opened a year ago by AndreaChiChengdu
  1

• 64位linux系统pip安装qwen_cpp报错，不支持？

  #32 opened a year ago by qianliyx
  0

• pip 安装 qwen-cpp 需要X86-32位的系统吗？能否支持X86-64位的系统

  #31 opened a year ago by zhougekaibenchi
  0

• Inferential capability of qwen.cpp for Qwen-14b-chat is different compared with Qwen-14b-chat of CUDA

  #30 opened a year ago by wertyac
  0

• Can you add an additional function to let convert.py support Qwen/Qwen-7B-Chat-Int4?

  #28 opened a year ago by x1ngzai
  0