OpenBMB/MiniCPM-V

Issues

• VisionEncoder里面的vit，用的是idefics2还是 hf4m那个啊？

  #94 opened 4 days ago by MonolithFoundation
  5

• 💡 [REQUEST] - Apple Silicon Mac GPU Support

  #146 opened 4 days ago by xinmingzh
  0

• [BUG] 多卡微调报错

  #144 opened 5 days ago by Zmeo
  1

• Where can I download the model file manually?

  #138 opened 6 days ago by goofusuper
  3

• [BUG] <mini cpm V2.0 finetune需要这么大显存么？>

  #145 opened 5 days ago by rover5056
  1

• ollama支持吗

  #97 opened 9 days ago by catzqaz
  21

• GGUF versions doesn't seem to run on llama.cpp (through LocalAI)

  #114 opened 7 days ago by naifmeh
  12

• Mac推理报错

  #96 opened 5 days ago by bookandlover
  3

• int4量化的精度下降了吗，下降多少？

  #143 opened 5 days ago by wodsoe
  3

• web_demo_2.5.py 调用 MiniCPM-Llama3-V-2_5-int4 模型报错

  #90 opened 8 days ago by triumph
  9

• 加上LLM_TYPE后报错ValueError

  #142 opened 5 days ago by Zmeo
  1

• int4和bffloat16推理时间问题（着急）

  #121 opened 5 days ago by githublsk
  3

• MiniCPM-V-2微调报错

  #131 opened 6 days ago by Zmeo
  3

• Fine-tuning results are not satisfactory

  #141 opened 5 days ago by Alxemade
  2

• 请问这个int4量化是怎么做的了，有相关的参考文件吗。

  #139 opened 5 days ago by Alxemade
  2

• MMMU val evaluation

  #140 opened 5 days ago by LongIslandWithoutIceTea
  1

• finetune/dataset.py | TypeError: Cannot cast array data from dtype('float64') to dtype('int32') according to the rule 'same_kind'

  #113 opened 8 days ago by JamePeng
  8

• MiniCPM-LLaMA3-V2.5 fine-tuning error

  #128 opened 6 days ago by yipoh
  4

• I want to know the use of the gpu of MiniCPM-Llama3-V 2.5

  #104 opened 6 days ago by praymich
  3

• Finetuned MiniCPM-V2.5 erro

  #93 opened 5 days ago by zfy1041264242
  4

• Can u opensoure the Chinese ocr part data?

  #98 opened 5 days ago by lucasjinreal
  3

• 多卡推理报错

  #125 opened 6 days ago by morgan-bc
  1

• when can we get the code of train from scratch?

  #105 opened 6 days ago by sankexin
  2

• finetune

  #135 opened 6 days ago by hn18001
  2

• 似乎扫描件效果不佳

  #136 opened 6 days ago by 0sengseng0
  1

• Expected all tensors to be on the same device, but found at least two devices

  #100 opened 6 days ago by CanvaChen
  2

• 多轮对话效果不佳

  #101 opened 6 days ago by CanvaChen
  4

• Deploy MiniCPM-V 2.5 with vllm

  #107 opened 6 days ago by zimoqingfeng
  2

• mlc-MiniCPM：回答异常

  #109 opened 6 days ago by zjts
  1

• minicpmv_example.py 文件不存在

  #110 opened 6 days ago by ipv6ok
  2

• RuntimeError: shape mismatch: value tensor of shape [1037] cannot be broadcast to indexing result of shape [1036]

  #116 opened 6 days ago by hujunchao
  2

• Questions about resource configuration

  #117 opened 6 days ago by uuhc
  4

• 请问能不能训练特定场景，然后转成ONNX

  #118 opened 6 days ago by shenyunhuan
  2

• 是否有，仿OpenAI API风格的demo运行文件

  #119 opened 6 days ago by xiaoToby
  3

• Model gets language confused - running gradio 2.5 demo.

  #124 opened 6 days ago by clearsitedesigns
  2

• 目前有计划ios的部署吗？

  #134 opened 6 days ago by Micla-SHL
  3

• 本地运行Int4版本报错

  #132 opened 6 days ago by ysyx2008
  2

• 图片无法上传。系统报错权限被拒绝。

  #133 opened 6 days ago by ysyx2008
  3

• v2-5 lora微调

  #130 opened 6 days ago by winsonxxxx
  2

• tech report or docs

  #122 opened 6 days ago by bpwl0121
  1

• package versions have conflicting dependencies.

  #120 opened 6 days ago by cvmilesight
  2

• M1,16G带不动如何卸载？

  #115 opened 7 days ago by ottngzy
  3

• conflicting dependencies，pip install -r requirements.txt 依赖项版本冲突问题。

  #127 opened 6 days ago by ysyx2008
  0

• doubt

  #123 opened 6 days ago by Sakshat10
  1

• 4张各12G的显卡仍然CUDA out of memory

  #102 opened 8 days ago by fishfree
  4

• Offical Demo web site error.

  #99 opened 8 days ago by wikeeyang
  2

• 支持ollama部署，thx

  #106 opened 8 days ago by ljch2018
  6

• 请问v2.0或者2.5可以做开放目标检测吗

  #103 opened 8 days ago by aliencaocao
  2

• BnB 8bit 在线量化报错

  #95 opened 9 days ago by QwertyJack
  2

• 微调数据是否支持多张图片作为输入？

  #91 opened 9 days ago by Crystalxd
  2