Will It Run AI
fal

AuraFlow v0.3

by fal

Open-source DiT model from fal.ai combining MMDiT and single DiT blocks in a Flux-like hybrid architecture. 6.35B transformer with T5-XL (~1.2B) text encoder. Apache 2.0 licensed — fully open for commercial use.

VRAM requirements, GPU fit, and setup notes for AuraFlow v0.3, including 8GB/12GB fit guidance where relevant. Recommended runtimes: ComfyUI and Diffusers support. Best download size: ~12.7 GB at FP16.

  • Apache 2.0 — fully open
  • Hybrid MMDiT + single DiT (Flux-like)
  • Up to 1536px
HuggingFaceDocumentation
949 downloads134 likes
ComfyUI, DiffusersFP16 safetensors

Your hardware

Detecting...

Parameters6.35B
Max Resolution1536×1536
Default Steps50
ArchitectureDIT
Licenseapache-2.0

Image Quality Benchmarks

Measured quality metrics for AuraFlow v0.3 outputs.

Human Preference Score65%

How often humans prefer this model's output (0-100%)

Aesthetic Score6.5

Visual quality and composition rating (5-9 scale)

CLIP Score0.27

Text-image alignment accuracy (higher is better)

VRAM Requirements by Resolution and Precision

Compare which GPUs can run AuraFlow v0.3 at different precisions. FP8 uses less memory than FP16 when available, and the grade shows how comfortably each GPU handles the workload.

FP16 (full precision)

ResolutionVRAM RequiredRTX 4090 24GBRTX 3060 12GBRTX 4060 8GBMacBook Pro M4 Pro 24GB
512×51216.5 GBSDFB
768×76816.7 GBSDFB
1024×102417.0 GBSFFB

Optimization Tips

Turbo / LCM distillation

Use distilled scheduler at 4-8 steps for faster iteration

Run with Python

Run with Python (diffusers)
from diffusers import AuraFlowPipeline
import torch

pipe = AuraFlowPipeline.from_pretrained(
    "fal/AuraFlow-v0.3",
    torch_dtype=torch.float16
)
pipe.to("cuda")

image = pipe(
    prompt="your prompt here",
    num_inference_steps=50,
    guidance_scale=3.5,
    height=1536,
    width=1536,
).images[0]
image.save("output.png")

Get started

Setup instructions for running AuraFlow v0.3 locally

1. Download the model

Get the checkpoint from HuggingFace

2. Place in:

ComfyUI/models/checkpoints/

3. Launch ComfyUI

python main.py

Memory Breakdown

VRAM allocation at 1024×1024 on RTX 4090 24GB (24 GB)

Required: 17.0 GBAvailable: 24.0 GB
Weights12.7 GB
VAE0.2 GB
Text Encoder2.4 GB
Activations0.6 GB
Overhead0.5 GB

Estimated Generation Time

Time per image at 1024×1024, 28 steps, FP16.

RTX 4090 24GB~18s
RTX 3060 12GB~1m 8s
RTX 4060 8GB~1m 42s
MacBook Pro M4 Pro 24GB~2m 26s

Sample Outputs

Available Formats, Downloads & Setup

Download AuraFlow v0.3 in the precision that matches your GPU. Lower precision usually means less VRAM pressure, while higher precision keeps more quality.

格式精度大小提供商
safetensors推荐FP1612.7 GBofficial下载

Related Workflows

You might also like

Frequently asked questions

FAQ — AuraFlow v0.3 VRAM, Runtimes & Fit

How much VRAM does AuraFlow v0.3 need?

AuraFlow v0.3 (6.35B parameters) requires approximately 17.0 GB of VRAM at FP16 precision for standard 1024×1024 image generation. If you want a lighter setup, lower precisions like FP8 can reduce memory use when available.

Can I run AuraFlow v0.3 on an 8GB GPU?

AuraFlow v0.3 usually needs more than 8GB for comfortable local use. Check the VRAM table above for the exact resolution and precision trade-off.

Does AuraFlow v0.3 work in ComfyUI and Diffusers?

AuraFlow v0.3 is marked for ComfyUI and Diffusers support in our catalog, so those are the runtimes we recommend first for local setup. If your workflow uses another front end, check the model's available formats and workflow notes above before downloading.

Can I run AuraFlow v0.3 on RTX 4090?

Yes, the RTX 4090 (24 GB VRAM) can run AuraFlow v0.3 comfortably at FP16. Expected generation time is around ~18s per image at 1024×1024.

Does AuraFlow v0.3 support ControlNet?

There are currently no known ControlNet adapters for AuraFlow v0.3. Check Hugging Face and Civitai for community-contributed adapters.

Does AuraFlow v0.3 have LoRA support?

No LoRA ecosystem. AuraFlow is relatively niche compared to Flux and SDXL. The LoRA ecosystem for AuraFlow v0.3 is rated as "none". Each LoRA adds roughly 0.0 GB of extra VRAM.

How fast is AuraFlow v0.3?

On a reference GPU (RTX 4090 24GB), AuraFlow v0.3 generates a 1024×1024 image in approximately ~18s at FP16 with 28 inference steps. Faster GPUs with higher memory bandwidth will produce images more quickly.

About AuraFlow v0.3

Use cases
artphotorealistic
Recommended runtimes
comfyuidiffusers

See also