search

Alpha-VLLM / Lumina-mGPT

Official Implementation of "Lumina-mGPT: Illuminate Flexible Photorealistic Text-to-Image Generation with Multimodal Generative Pretraining"
https://arxiv.org/abs/2408.02657
452 stars 19 forks source link

Has the lumina-mGPT done any quantitative evaluation on generation or VLM benchmark? #7

Open muxizju opened 1 month ago

muxizju commented 1 month ago

like generation FID，or the VLM benchmark, e.g. mme vqa

gaopengpjlab commented 1 month ago

to be done