AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

SANA-Sprint: One-Step Diffusion with Continuous-Time Consistency Distillation

Junsong Chen, Shuchen Xue, Yuyang Zhao, Jincheng Yu, Sayak Paul, Junyu Chen, Han Cai, Song Han, Enze Xie

ICCV 2025 · Image generation › Fast diffusion sampling · approach: Fast diffusion ODE solvers · 66 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add LADD adversarial loss to sCM (ablation)SANA-Sprint · MJHQ-30K 2-stepFID: 8.93 -> 8.11Improved
add LADD adversarial loss to sCM (ablation)SANA-Sprint · MJHQ-30K 2-stepCLIP score: 27.51 -> 28.02Improved
use LADD alone instead of sCM plus LADD (ablation)SANA-Sprint · MJHQ-30K 2-stepFID: 8.11 -> 12.20Hurt
add CFG embedding to student (ablation)SANA-Sprint · MJHQ-30K 2-stepCLIP score: 27.15 -> 28.09Improved
add CFG embedding to student (ablation)SANA-Sprint · MJHQ-30K 2-stepFID: 9.23 -> 8.72Improved
add max-time weighting at t=pi/2 (50%) (ablation)SANA-Sprint · MJHQ-30K 2-stepFID: 9.44 -> 8.32Improved
raise max-time probability from 50% to 70% (ablation)SANA-Sprint · MJHQ-30K 2-stepFID: 8.32 -> 8.11No clear effect
sCM:LADD weight 1.0:0.5 vs 1.0:1.0 (ablation)SANA-Sprint · MJHQ-30K 2-stepFID: 8.81 -> 8.43Improved
sCM:LADD weight 1.0:0.1 vs 1.0:0.5 (ablation)SANA-Sprint · MJHQ-30K 2-stepFID: 8.43 -> 8.90Hurt
training-free flow to TrigFlow transformationSANA · MJHQ-30K 50-step EulerFID: 5.81 -> 5.73No clear effect
use schedule transfer when fine-tuning flow model to TrigFlow (ablation)SANA · fine-tuningtraining stabilityImproved
set c_noise(t)=t instead of 1000t (dense time embedding) (ablation)SANA teacher · gradient normgradient norm: exceeding 10^3 -> more stable levelsImproved

Later papers that compared with this method

2 later accepted papers report SANA-Sprint 0.6B as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy