AuraFlow

Released
2024-07-12
Last refreshed
2026-05-19
Status
Researched 174d ago
Open sourceCommercial use: permittedMultimodal

AuraFlow is a released general LLM work model with open-source; evaluate it while provider pricing coverage matures.

Specifications
Family
AuraFlow
Released
2024-07-12
Parameters
6.8B
Architecture
Diffusion Transformer
Specialization
image-generation
Openness
Open source
License
Apache 2.0OSI-approvedPermitted
Weights
Available
Code
Unknown
Created by
fal

Blazing-fast AI inference technology

Oakland, California, United States
Founded 2020
Website
Pricing

No tracked provider pricing is available yet.

Links
Website HuggingFace

About

AuraFlow by fal.ai. At 6.8B parameters, the largest open-source flow-based text-to-image model at launch. Uses flow matching (direct noise-to-image mapping) rather than traditional diffusion, achieving strong semantic coherence for complex multi-object compositions. Released July 2024 under Apache 2.0. Available on HuggingFace (fal/AuraFlow) and Fal API (fal-ai/aura-flow).

Providers

No active provider routes are tracked for this model yet.

Capabilities

Multimodal

Benchmark peer barsfor Coding

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.