search

DLYuanGod / TinyGPT-V

TinyGPT-V: Efficient Multimodal Large Language Model via Small Backbones
BSD 3-Clause "New" or "Revised" License
1.24k stars 76 forks source link

How long does it take to train? #3

Open zxti opened 10 months ago

zxti commented 10 months ago

Thank you for this model, excited to try training!

I didn't see this in the paper (sorry if I missed it), but how long does it take to train? Any numbers you already have would be helpful to know what to expect—whether it's all stages / specific stages, or single 28GB GPU / multiple, etc. Thank you!

DLYuanGod commented 10 months ago

We use a single 3090 GPU(24G).

Stage 1 About 8 hours of training, Stage 2 About 4 hours, Stage 3 About 20 mins, Stage 4 About 8 hours or more.

Currently Stage 4 is still in a testing state, as it is not actually performing well (although it is able to perform well on the assessment results), and it is recommended that you only study up to Stage 3.

We are actively working on updating our model and expect to release a more robust version within two weeks!

Thank you for your interest.