# UltraFlux: Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect Ratios

arXiv: https://arxiv.org/abs/2511.18050 · HTML: https://mlsanity.com/paper/2511.18050/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 38 of its 43 references that are in the map.
- First posted on arXiv 22 Nov 2025.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 30 of 38)

- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [MUSIQ: Multi-scale Image Quality Transformer](https://mlsanity.com/paper/2108.05997/) (2021)
- [Improved Techniques for Training Score-Based Generative Models](https://mlsanity.com/paper/2006.09011/) (2020)
- [FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space](https://mlsanity.com/paper/2506.15742/) (2025)
- [Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2305.01569/) (2023)
- [YaRN: Efficient Context Window Extension of Large Language Models](https://mlsanity.com/paper/2309.00071/) (2023)
- [MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generation](https://mlsanity.com/paper/2302.08113/) (2023)
- [Q-Align: Teaching LMMs for Visual Scoring via Discrete Text-Defined Levels](https://mlsanity.com/paper/2312.17090/) (2023)
- [SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformers](https://mlsanity.com/paper/2410.10629/) (2024)
- [PixArt-Sigma: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation](https://mlsanity.com/paper/2403.04692/) (2024)
- [Long-CLIP: Unlocking the Long-Text Capability of CLIP](https://mlsanity.com/paper/2403.15378/) (2024)
- [Seedream 4.0: Toward Next-generation Multimodal Image Generation](https://mlsanity.com/paper/2509.20427/) (2025)
- [Efficient Diffusion Training via Min-SNR Weighting Strategy](https://mlsanity.com/paper/2303.09556/) (2023)
- [HPSv3: Towards Wide-Spectrum Human Preference Score](https://mlsanity.com/paper/2508.03789/) (2025)
- [ScaleCrafter: Tuning-free Higher-Resolution Visual Generation with Diffusion Models](https://mlsanity.com/paper/2310.07702/) (2023)
- [Aesthetic Post-Training Diffusion Models from Generic Preferences with Step-by-step Preference Optimization](https://mlsanity.com/paper/2406.04314/) (2024)
- [Diffusion-4K: Ultra-High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2503.18352/) (2025)
- [FouriScale: A Frequency Perspective on Training-Free High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.12963/) (2024)
- [Make a Cheap Scaling: A Self-Cascade Diffusion Model for Higher-Resolution Adaptation](https://mlsanity.com/paper/2402.10491/) (2024)
- [ArtiMuse: Fine-Grained Image Aesthetics Assessment with Joint Scoring and Expert-Level Understanding](https://mlsanity.com/paper/2507.14533/) (2025)
- [ElasticDiffusion: Training-free Arbitrary Size Image Generation through Global-Local Content Separation](https://mlsanity.com/paper/2311.18822/) (2023)
- [HiDiffusion: Unlocking Higher-Resolution Creativity and Efficiency in Pretrained Diffusion Models](https://mlsanity.com/paper/2311.17528/) (2023)
- [CLEAR: Conv-Like Linearization Revs Pre-Trained Diffusion Transformers Up](https://mlsanity.com/paper/2412.16112/) (2024)
- [Resonance RoPE: Improving Context Length Generalization of Large Language Models](https://mlsanity.com/paper/2403.00071/) (2024)
- [Denoising Task Difficulty-based Curriculum for Training Diffusion Models](https://mlsanity.com/paper/2403.10348/) (2024)
- [OmniSync: Towards Universal Lip Synchronization via Diffusion Transformers](https://mlsanity.com/paper/2505.21448/) (2025)

## Built on by (most cited first, 6 of 6)

- [SANA-Streaming: Real-time Streaming Video Editing with Hybrid Diffusion Transformer](https://mlsanity.com/paper/2605.30409/) (2026)
- [PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion](https://mlsanity.com/paper/2605.23902/) (2026)
- [OP4KSR: One-Step Patch-Free 4K Super-Resolution with Periodic Artifact Suppression](https://mlsanity.com/paper/2605.13457/) (2026)
- [PixelWizard: Towards Efficient High-Fidelity Video Generation at Ultra-Large Spatial Resolution](https://mlsanity.com/paper/2605.25801/) (2026)
- [PixVerve: Advancing Native UHR Image Generation to 100MP with a Large-Scale High-Quality Dataset](https://mlsanity.com/paper/2605.20147/) (2026)
- [EDITBRIDGE: Towards Faithful and Efficient Ultra-High-Resolution Image Editing](https://mlsanity.com/paper/2608.18063/) (2026)

## Abstract (arXiv)

Diffusion transformers have recently delivered strong text-to-image generation around 1K resolution, but we show that extending them to native 4K across diverse aspect ratios exposes a tightly coupled failure mode spanning positional encoding, VAE compression, and optimization. Tackling any of these factors in isolation leaves substantial quality on the table. We therefore take a data-model co-design view and introduce UltraFlux, a Flux-based DiT trained natively at 4K on MultiAspect-4K-1M, a 1M-image 4K corpus with controlled multi-AR coverage, bilingual captions, and rich VLM/IQA metadata for resolution- and AR-aware sampling. On the model side, UltraFlux couples (i) Resonance 2D RoPE with YaRN for training-window-, frequency-, and AR-aware positional encoding at 4K; (ii) a simple, non-adversarial VAE post-training scheme that improves 4K reconstruction fidelity; (iii) an SNR-Aware Huber Wavelet objective that rebalances gradients across timesteps and frequency bands; and (iv) a Stage-wise Aesthetic Curriculum Learning strategy that concentrates high-aesthetic supervision on high-noise steps governed by the model prior. Together, these components yield a stable, detail-preserving 4K DiT that generalizes across wide, square, and tall ARs. On the Aesthetic-Eval at 4096 benchmark and multi-AR 4K settings, UltraFlux consistently outperforms strong open-source baselines across fidelity, aesthetic, and alignment metrics, and-with a LLM prompt refiner-matches or surpasses the proprietary Seedream 4.0.