Shadowgraph
TrendingBuzzResearch
Chrome extension
Sign in

kooroshrh/AReA

Space

AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution Diffusion Transformers

5

21 commits

1 linked in READMEs

updated Oct 15, 2026

See the code

README

AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution Diffusion Transformers

Official demo of AReA, a training-free, model-agnostic attention mechanism that lets diffusion transformers efficiently generate images beyond their training resolution, with no fine-tuning, extra parameters or RoPE extrapolation.

Due to limited GPU resources, this demo runs FLUX.1 only. For other models, see the GitHub repository.

AReA samples

diffusion-transformer
flux
gradio
high-resolution
text-to-image

Similar repos

KooroshRH/AReA

Official implementation of "AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution…

36

GuoLanqing/Awesome-High-Resolution-Diffusion

🔥🔥🔥A curated list of papers on recent diffusion-based high-resolution image and video synthesis…

170

Owen777/UltraFlux-v1

Model

UltraFlux: Data–Model Co-Design for Native 4K, Multi-AR Text-to-Image

210

Owen777/UltraFlux-v1-1-Transformer

Model

UltraFlux: Data–Model Co-Design for Native 4K, Multi-AR Text-to-Image

4

Lakonik/AsymFLUX.2-klein

Space

Official demo of the paper:

43

jiaxiangc/res-adapter

Model

ResAdapter Model Card

54

bytedance/res-adapter

[AAAI 2025] Official codes of "ResAdapter: Domain Consistent Resolution Adapter for Diffusion…

754

OAOA/InvSR

Space

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

473

Linked in READMEs

KooroshRH/AReA

Official implementation of "AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution…

36

README badge

kooroshrh/AReA

Space

AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution Diffusion Transformers

5

21 commits

1 linked in READMEs

updated Oct 15, 2026

See the code

README

AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution Diffusion Transformers

Official demo of AReA, a training-free, model-agnostic attention mechanism that lets diffusion transformers efficiently generate images beyond their training resolution, with no fine-tuning, extra parameters or RoPE extrapolation.

Due to limited GPU resources, this demo runs FLUX.1 only. For other models, see the GitHub repository.

AReA samples

diffusion-transformer
flux
gradio
high-resolution
text-to-image

Similar repos

KooroshRH/AReA

Official implementation of "AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution…

36

GuoLanqing/Awesome-High-Resolution-Diffusion

🔥🔥🔥A curated list of papers on recent diffusion-based high-resolution image and video synthesis…

170

Owen777/UltraFlux-v1

Model

UltraFlux: Data–Model Co-Design for Native 4K, Multi-AR Text-to-Image

210

Owen777/UltraFlux-v1-1-Transformer

Model

UltraFlux: Data–Model Co-Design for Native 4K, Multi-AR Text-to-Image

4

Lakonik/AsymFLUX.2-klein

Space

Official demo of the paper:

43

jiaxiangc/res-adapter

Model

ResAdapter Model Card

54

bytedance/res-adapter

[AAAI 2025] Official codes of "ResAdapter: Domain Consistent Resolution Adapter for Diffusion…

754

OAOA/InvSR

Space

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

473

Linked in READMEs

KooroshRH/AReA

Official implementation of "AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution…

36

README badge