AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

StreamFlow: Theory, Algorithm, and Implementation for High-Efficiency Rectified Flow Generation

Sen Fang, Hongbin Zhong, Yalin Feng, Yanxin Zhang, Dimitris N. Metaxas

ICML 2026 · Image generation › Fast diffusion sampling · approach: Rectified flow straightening · 0 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 xformer architecture and taesd decoder (ablation)Vanilla pipeline · Ablation, FPSFPS: 1.87 -> 2.3Improved
add VFB (batched velocity field) with taesd (ablation)xformer+taesd · Ablation, FPSFPS: 2.3 -> 8.0Improved
add TensorRT compilation (ablation)xformer+taesd+VFB · Ablation, FPSFPS: 8.0 -> 10.05Improved
add compatibility layer (ablation)tensorrt+taesd+VFB · Ablation, FPSFPS: 10.05 -> 9.8Hurt
use 8 inference steps instead of 4 (ablation)Optimized pipeline · Ablation, FPSFPS: 4.5 FPS vs 2.22 FPS 4-step baselineHurt
use taesd instead of original VAE (ablation)xformer+VFB · Quality study, text-to-imageCLIP score: 96.74 -> 97.26Improved
use taesd instead of original VAE (ablation)xformer+VFB · Quality study, text-to-imageFID: 31.82 -> 30.75Improved
use taesd instead of original VAE (ablation)tensorrt+VFB · Quality study, text-to-imageCLIP score: 96.08 -> 96.25Improved
use taesd instead of original VAE (ablation)tensorrt+VFB · Quality study, text-to-imageFID: 31.16 -> 31.28No clear effect
apply StreamFlow acceleration (xformer+taesd+VFB)Baseline Vanilla · Quality study, text-to-imageFID: 29.44 -> 30.75No clear effect
apply tensorrt+taesd+VFBBaseline Vanilla · Power/Memory comparison, image generationpower (W): 205.28 -> 110.51Improved
apply tensorrt+taesd+VFBBaseline Vanilla · Power/Memory comparison, image generationFPS: 1.87 -> 10.05Improved

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