Video-Text-to-Text
Transformers
Safetensors
English
llava
text-generation
multimodal
Eval Results
Inference Endpoints

Difference between 7B-DPO and 7B-Qwen2

#7
by RachelZhou - opened

Hi, I have a question regarding the differences between the following models:

  • 7B
  • 7B-Qwen2
  • 7B-DPO

After the recent update, it seems unclear whether the 7B and 7B-DPO models still exist. Since I’m conducting studies with the 7B-DPO model, could you please clarify the differences between these models or point me to any resources that outline these differences?

LMMs-Lab org

Hi,the 7B-Qwen2 is our latest model

  1. Trained on the newest data
  2. Using Qwen2 as backbone
  3. without DPO stage

Sign up or log in to comment