SANA-Sprint: One-Step Diffusion with Continuous-Time Consistency Distillation
Authors:
- Jincheng YuJYJincheng Yu
- Song HanSHSong Han
Preprint
Abstract
This paper presents SANA-Sprint, an efficient diffusion model for ultra-fast text-to-image (T2I) generation. SANA-Sprint is built on a pre-trained foundation model and augmented with hybrid distillation, dramatically reducing inference steps from 20 to 1-4. We introduce three key innovations: (1) We propose a training-free approach that transforms a pre-trained flow-matching model for continuous-time consistency distillation (sCM), eliminating costly training from scratch and achieving high training efficiency. Our hybrid distillation strategy combines sCM with latent adversarial distillation (LADD): sCM ensures alignment with the teacher model, while LADD enhances single-step generation fidelity. (2) SANA-Sprint is a unified step-adaptive model that achieves high-quality generation in 1-4 steps, eliminating step-specific training and improving efficiency. (3) We integrate ControlNet with SANA-Sprint for real-time interactive image generation, enabling instant visual feedback for user interaction. SANA-Sprint establishes a new Pareto frontier in speed-quality tradeoffs, achieving state-of-the-art performance with 7.59 FID and 0.74 GenEval in only 1 step - outperforming FLUX-schnell (7.94 FID / 0.71 GenEval) while being 10x faster (0.1s vs 1.1s on H100). It also achieves 0.1s (T2I) and 0.25s (ControlNet) latency for 1024 x 1024 images on H100, and 0.31s (T2I) on an RTX 4090, showcasing its exceptional efficiency and potential for AI-powered consumer applications (AIPC). Code and pre-trained models will be open-sourced.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- SnapGen: Taming High-Resolution Text-To-Image Models for Mobile Devices with Efficient Architectures and Training2025
- DiG: Scalable and Efficient Diffusion Models with Gated Linear Attention2025
- PIXART-$\Sigma $: Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation2024
- PaGoDA: Progressive Growing of a One-Step Generator from a Low-Resolution Diffusion Teacher2024
- Scalable Diffusion Models with Transformers2023
- Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models2023
- ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation2023
- GenEval: An object-focused framework for evaluating text-to-image alignment2023
- SnapFusion: Text-to-Image Diffusion Model on Mobile Devices within Two Seconds2023
- A Connection Between Score Matching and Denoising AutoencodersNeural Computation, 2011