AuraFlow v0.x series, to date the largest (6.8B) and highest fidelity (0.7+ on GenEval) open sourced text to image model.
fal
Enterprise
company
AI & ML interests
generative media platform for developers
Organization Card
Hi there! We are fal, a generative media platform for developers running open source image/video/audio/3D models in the fastest way possible.
Check out our model gallery and build using ready-to-use APIs.
Or visit our docs to build custom applications on our Serverless GPUs.
models
10
fal/AuraEquiVAE
Updated
•
15
fal/AuraFace-v1
Updated
•
67
fal/AuraFlow-v0.3
Text-to-Image
•
Updated
•
1.06k
•
105
fal/AuraSR-v2
Image-to-Image
•
Updated
•
4.21k
•
219
fal/AuraFlow-v0.2
Text-to-Image
•
Updated
•
704
•
144
fal/AuraFlow
Text-to-Image
•
Updated
•
763
•
621
fal/AuraSR
Updated
•
880
•
294
fal/moondream2-docci-instruct
Image-Text-to-Text
•
Updated
•
26
•
9
fal/resources
Updated
•
1
fal/teed
Updated