swift
swift copied to clipboard
VRAM requirement for full sft deepseek VL 7B
Describe the bug How much VRAM is needed to finetune the 7b VL model?
# Experimental Environment: A100
# GPU Memory Requirement: 80GB
# Runtime: 2.5 hours
CUDA_VISIBLE_DEVICES=0 \
swift sft \
--model_type qwen1half-7b-chat \
--dataset blossom-math-zh \
--num_train_epochs 5 \
--sft_type full \
--output_dir output \
--eval_steps 500 \
The docs say one needs 80GB for a normal 7b model, however when I try to train the DeepSeek VL 7B on the research rig with an A100 I get an OOM. When trying to split across 4 GPUs (1 A100 and 3 4090s), it does not utilize the A100 and OOMs with the 3 4090s before training can start