Flexible 1D tokenizers and VAEs from https://flextok.epfl.ch/
EPFL VILAB
AI & ML interests
None defined yet.
Recent Activity
Collections
4
spaces
4
models
34
EPFL-VILAB/flextok_d18_d28_dfn
Updated
•
3.94k
EPFL-VILAB/flextok_d18_d28_in1k
Updated
•
9
EPFL-VILAB/flextok_d18_d18_in1k
Updated
EPFL-VILAB/flextok_d12_d12_in1k
Updated
•
169
EPFL-VILAB/flextok_vae_c16
Updated
•
6
EPFL-VILAB/flextok_vae_c8
Updated
•
6
EPFL-VILAB/flextok_vae_c4
Updated
•
6
EPFL-VILAB/VPE-ViPer
Image-Text-to-Text
•
Updated
•
34
•
2
EPFL-VILAB/Metric-ViPer
Image-Text-to-Text
•
Updated
•
75
•
2
EPFL-VILAB/4M-21_XL
Any-to-Any
•
Updated
•
116
•
35
datasets
None public yet