mlsanityData to 7 Oct 2026

NanoFLUX: Distillation-Driven Compression of Large Text-to-Image Generation Models for Mobile Devices

arXiv 2602.06879 ↗First posted 6 Feb 2026cs.CV cs.AIcited by 0

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

While large-scale text-to-image diffusion models continue to improve in visual quality, their increasing scale has widened the gap between state-of-the-art models and on-device solutions. To address this gap, we introduce NanoFLUX, a 2.4B text-to-image flow-matching model distilled from 17B FLUX.1-Schnell using a progressive compression pipeline designed to preserve generation quality. Our contributions include: (1) A model compression strategy driven by pruning redundant components in the diffusion transformer, reducing its size from 12B to 2B; (2) A ResNet-based token downsampling mechanism that reduces latency by allowing intermediate blocks to operate on lower-resolution tokens while preserving high-resolution processing elsewhere; (3) A novel text encoder distillation approach that leverages visual signals from early layers of the denoiser during sampling. Empirically, NanoFLUX generates 512 x 512 images in approximately 2.5 seconds on mobile devices, demonstrating the feasibility of high-quality on-device text-to-image generation.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 37 of its 47 references that are in the map. First posted on arXiv 6 Feb 2026.

Builds on

37 of 47 references.

Closest ancestors · 7

Same direction, newest first

  1. 2025Scaling Down Text Encoders of Text-to-Image Diffusion Models25 Mar
  2. 2025SANA-Sprint: One-Step Diffusion with Continuous-Time Consistency Distillation12 Mar
  3. 2024Efficient Pruning of Text-to-Image Models: Insights from Pruning Stable Diffusion22 Nov
  4. 2024ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment8 Mar
  5. 2023MobileDiffusion: Instant Text-to-Image Generation on Mobile Devices28 Nov
  6. 2023GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignment17 Oct
  7. 2023SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Seconds1 Jun

Other references · 30

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  2. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  3. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
  4. 2016Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Networkcited by 1,199
  5. 2023Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesiscited by 847
  6. 2025Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformercited by 247
  7. 2023Token Merging for Fast Stable Diffusioncited by 220
  8. 2019Understanding and Improving Layer Normalizationcited by 213
  9. 2023DepGraph: Towards Any Structural Pruningcited by 202
  10. 2025HPSv3: Towards Wide-Spectrum Human Preference Scorecited by 196
  11. 2021Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesiscited by 140
  12. 2024FORA: Fast-Forward Caching in Diffusion Transformer Accelerationcited by 127
18 more
  1. 2024DiTFastAttn: Attention Compression for Diffusion Transformer Modelscited by 109
  2. 2025OneIG-Bench: Omni-dimensional Nuanced Evaluation for Image Generationcited by 81
  3. 2024LD-Pruner: Efficient Pruning of Latent Diffusion Models using Task-Agnostic Insightscited by 65
  4. 2024U-DiTs: Downsample Tokens in U-Shaped Diffusion Transformerscited by 51
  5. 2024TinyFusion: Diffusion Transformers Learned Shallowcited by 44
  6. 2024Rethinking Token-wise Feature Caching: Accelerating Diffusion Transformers with Dual Feature Cachingcited by 26
  7. 2024Layer- and Timestep-Adaptive Differentiable Token Compression Ratios for Efficient Diffusion Transformerscited by 21
  8. 2024Importance-Based Token Merging for Efficient Image and Video Generationcited by 20
  9. 2025DiTFastAttnV2: Head-wise Attention Compression for Multi-Modality Diffusion Transformerscited by 19
  10. 2025ToMA: Token Merge with Attention for Diffusion Modelscited by 17
  11. 2025CAT Pruning: Cluster-Aware Token Pruning For Text-to-Image Diffusion Modelscited by 16
  12. 2025Neodragon: Mobile Video Generation using Diffusion Transformercited by 10
  13. 2025MosaicDiff: Training-free Structural Pruning for Diffusion Model Acceleration Reflecting Pretraining Dynamicscited by 9
  14. 2025HierarchicalPrune: Position-Aware Compression for Large-Scale Diffusion Modelscited by 8
  15. 2024LayerMerge: Neural Network Depth Compression through Layer Pruning and Mergingcited by 7
  16. 2025EDiT: Efficient Diffusion Transformers with Linear Compressed Attentioncited by 7
  17. 2025Pruning for Sparse Diffusion Models based on Gradient Flowcited by 5
  18. 2025FraQAT: Quantization Aware Training with Fractional bitscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025UltraFlux: Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect Ratios
  2. 2024MegaFusion: Extend Diffusion Models towards Higher-resolution Image Generation without Further Tuning
  3. 2024SnapGen: Taming High-Resolution Text-to-Image Models for Mobile Devices with Efficient Architectures and Training
  4. 2024EdgeFusion: On-Device Text-to-Image Generation
  5. 2025DiT-Air: Revisiting the Efficiency of Diffusion Model Architecture Design in Text to Image Generation
  6. 2026ElasticDiT: Efficient Diffusion Transformers via Elastic Architecture and Sparse Attention for High-Resolution Image Generation on Mobile Devices
  7. 2024Meissonic: Revitalizing Masked Generative Transformers for Efficient High-Resolution Text-to-Image Synthesis
  8. 2025MobileI2V: Fast and High-Resolution Image-to-Video on Mobile Devices