zhixiangwei
/

vlm1b-hqclip-xlarge-vitl14-clipa

Model card Files Files and versions Community

Pretrain CLIP-L-14 on VLM-1B using CLIPA.

Dataset	Performance
ImageNet 1k	0.78628
ImageNet V2	0.7132
ImageNet-A	0.662
ImageNet-O	0.4085
ImageNet-R	0.900967
ImageNet Sketch	0.673643
ObjectNet	0.716324
IN-shifts	0.679106
VTAB	0.604791
MSCOCO	0.580741
Flickr30k	0.841
WinoGAViL	0.509853
Retrieval	0.643865
Avg.	0.637904

license: apache-2.0

Downloads last month: 22

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Dataset used to train zhixiangwei/vlm1b-hqclip-xlarge-vitl14-clipa

Collection including zhixiangwei/vlm1b-hqclip-xlarge-vitl14-clipa

HQCLIP

7 items • Updated 13 days ago