Meituan-AutoML/MobileVLM

Issues

• what is in the pretrain_data for mobileVLM2? @er-muyue

  #52 opened a month ago by xiyuanhao
  0

• where can I download mm_projector

  #51 opened a month ago by wuwu-C
  0

• About the supply of mm_projector.bin

  #44 opened 2 months ago by Bobby-youngking
  1

• funtune with lora error

  #17 opened 5 months ago by BadUncleBoy
  8

• running gguf Mobile VLM on Mac (metal)

  #50 opened a month ago by rezacopol
  0

• llama.c转换mobileLVM-V2报错

  #29 opened 4 months ago by shinerdeng
  5

• sft loss curve

  #49 opened a month ago by shufangxun
  0

• vision_tower读取不到内容，返回none

  #5 opened 4 months ago by caramel678
  4

• Results of Evaluation with MobileVLM_V2-1.7B on GQA/POPE/MME Dataset is Different from that in Paper

  #41 opened 2 months ago by ZiangWu-77
  0

• 关于手机上运行时间的求教

  #13 opened 3 months ago by JennieGao-njust
  12

• Interest in training a Phi-3 version?

  #48 opened 2 months ago by kinchahoy
  0

• MobileVLM_V2-3B doesn't work on llama.cpp ?

  #39 opened 3 months ago by weizhou1991
  1

• pre-train dataset

  #47 opened 2 months ago by FabienCode
  0

• Can MobileVLM v2 receive multiple images at once？

  #37 opened 3 months ago by wuwu-C
  2

• I'm curious that why MobileLLaMA 2.7B is faster then OpenLLaMA reported in v1 paper?

  #46 opened 2 months ago by XiaohuJoshua
  0

• how to finetune on user's own data

  #11 opened 3 months ago by 77h2l
  5

• questions about fine tune

  #43 opened 2 months ago by Bobby-youngking
  0

• CUDA out of memory

  #38 opened 3 months ago by qazzombie
  3

• 有中文的readme没有？两张rtx A4000跑不起来，报错

  #32 opened 3 months ago by life2048
  2

• How to use our own dataset to test?

  #34 opened 3 months ago by QvQKing
  2

• 请问推理支持batch跑吗？

  #33 opened 3 months ago by LianghuiGuo
  1

• 请问支持量化推理吗？

  #36 opened 3 months ago by LianghuiGuo
  1

• A100推理为啥很慢呢？是我Inference姿势不对吗？

  #35 opened 3 months ago by LianghuiGuo
  2

• mobilevlm_v2 with lora

  #30 opened 3 months ago by rangmiao
  2

• Problem of Finetune

  #31 opened 3 months ago by QvQKing
  2

• Llama.cpp usage does not work with the latest v2-1.7B model

  #26 opened 3 months ago by l3utterfly
  3

• Hugging Face demo

  #24 opened 3 months ago by merveenoyan
  2

• clip部分 ldp 跟 直接编码mlp的时间对比

  #23 opened 3 months ago by JennieGao-njust
  3

• my linux workstation die when I run infernece.py

  #3 opened 3 months ago by ssifood
  5

• Warning about tokenizer

  #42 opened 3 months ago by xushilin1
  0

• Repetitive response problem with VLM

  #27 opened 4 months ago by Muhammad4hmed
  11

• FlashAttention1.0.9 error in 2080Ti

  #40 opened 3 months ago by miraclewkf
  0

• any lora for stage 2?

  #22 opened 4 months ago by shawnricecake
  2

• Is it possible to provide pretrained mm_projector.bin

  #19 opened 4 months ago by KP-Zhang
  2

• Where can I obtain the data used for MobileVLM v2 training?

  #28 opened 4 months ago by xushilin1
  2

• Some question on selection device and card

  #15 opened 4 months ago by alone4u1314
  3

• Question about mobile platform deploy

  #10 opened 4 months ago by JiaoYanMoGu
  2

• 你们是怎么在骁龙888上运行的？

  #8 opened 4 months ago by xsailor511
  3

• how to use lora

  #6 opened 4 months ago by liuweixue001
  9

• 实现移动端部署的路径

  #4 opened 4 months ago by JennieGao-njust
  4

• inference on iPhone

  #2 opened 4 months ago by tstanek390
  7

• 请问支持中文推理/中文数据微调吗？

  #20 opened 5 months ago by LianghuiGuo
  1

• Does it support Chinese instruction for mobileVLM?

  #16 opened 5 months ago by ghoshaw
  1

• Significant Performance Drop in Model Replication

  #18 opened 5 months ago by rangmiao
  6

• 'get_tensor' error occurs during inference on mobile devices

  #14 opened 5 months ago by wwqq
  2

• NotImplementedError: Cannot copy out of meta tensor; no data!

  #7 opened 5 months ago by 4ku
  4

• Benchmark Code

  #9 opened 5 months ago by iszhiyang
  4

• MobileVLM weights?

  #1 opened 5 months ago by pHaeusler
  3