AuraFlow v0.3
Parameters
6.8B
Modality
Image
Auxiliary Parameters
4.78B
Architecture
Diffusion Transformer
License
Apache 2.0
Release Date
25 Jul 2024
API Pricing (per 1M)
Self-hosted only
Hardware Sizing
Precision
Workload Memory: 6.10 GB
INT4 · 1024x1024
DiT Backbone Weights: 3.17 GB
VAE Decode Workspace: 0.44 GB
Runtime & Graphs: 2.50 GB
Minimum Required Hardware
1x NVIDIA L4 24GB
6.1 / 21.6 GB usable
1x RTX 3060 12GB
6.1 / 10.8 GB usable
Apple M4 24GB
6.1 / 18.0 GB usable
Determine requirements for your specific workload:
Open in VRAM Calculator
Architecture Diagram
About AuraFlow v0.3
AuraFlow v0.3 is an open-source 6.8B parameter rectified-flow diffusion transformer trained by fal.ai, featuring dual-stream attention layers with UMT5-XXL conditioning.
Technical Specifications
Attention
Attention Heads
12
Attention Head Dimension
256
Dimensions
Auxiliary Parameters
4.78B
Hidden Dimension Size
2,048
Number of Layers
28