OutOfMemoryError: CUDA out of memory. Tried to allocate 112.00 MiB. GPU

#51
by mdeniz1 - opened

Hi
I use the following code and I receive the error "OutOfMemoryError: CUDA out of memory. Tried to allocate 112.00 MiB. GPU"

I have nvidia 3050. what can be the problem?

import torch

from transformers import AutoProcessor, AutoModelForVision2Seq
from transformers.image_utils import load_image
DEVICE = "cuda:0"
from transformers import BitsAndBytesConfig
from transformers import AwqConfig

quantization_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_use_double_quant=True,
bnb_4bit_compute_dtype=torch.float16
)
model = AutoModelForVision2Seq.from_pretrained(
"HuggingFaceM4/idefics2-8b",
torch_dtype=torch.float16,
quantization_config=quantization_config,
).to(DEVICE)

hi @mdeniz1 it looks like you don't have enough gpu memory to load the model?

hi @mdeniz1 it looks like you don't have enough gpu memory to load the model?

that is correct! thank you

mdeniz1 changed discussion status to closed

Sign up or log in to comment