Meituan-AutoML / MobileVLM

Strong and Open Vision Language Assistant for Mobile Devices
Apache License 2.0
1.04k stars 66 forks source link

CUDA out of memory #38

Open qazzombie opened 8 months ago

qazzombie commented 8 months ago

我在A800上,设置batchsize=1,为啥还会oom呀

ghoshaw commented 8 months ago

同问,训练mobilellama的sft需要多大显存?

miraclewkf commented 8 months ago

我在A800上,设置batchsize=1,为啥还会oom呀

请问你解决问题了吗?

wyddmw commented 7 months ago

I met the same problem on a 40G A100 GPU. Waiting for feedback.