AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Otil: Accelerating Diffusion Model Inference via Communication-Efficient Multi-GPU Parallelism

Xin Li, Shujun Tian, Tao Lu, Han Bao, Zonghui Wang, Wenzhi Chen

CVPR 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration

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
transmit only 1/4 of latent sub-blocks instead of all (ablation)Otil · SDXL 1.0, 2 GPUsper-step latency (ms): 43.5324 -> 27.5324 (LPIPS 0.0139 -> 0.0142)Improved
transmit only 1/16 of sub-blocks (ablation)Otil · SDXL 1.0, 2 GPUsLPIPS: 0.0142 -> 0.1342Hurt
transmit only 1/4 sub-blocks instead of full on 4 GPUs (ablation)Otil · SDXL 1.0, 4 GPUsLPIPS: 0.0693 -> 0.1350Hurt
use 8x8 sub-blocks instead of 4x4 (ablation)Otil · SDXL 1.0, 2 GPUsLPIPS: 0.0571 -> 0.0142Improved
use 8x8 sub-blocks instead of 16x16 (ablation)Otil · SDXL 1.0, 2 GPUsLPIPS: 0.0294 -> 0.0142Improved
rank sub-blocks by cosine similarity instead of random, SSIM, mutual information or dHash (ablation)Otil · latent-to-pixel change alignment (AUC)AUCImproved
add dynamic polling so every sub-block is revisited cyclicallyOtil · COCO Captions 2014image qualityImproved
combine with few-step samplers (UniPC) and 2 GPUsOtil · SDXL 1.0, 30 stepsspeedup: 1.66 -> 2.84Improved
combine with LoRA SD 1.5 on 2 GPUsOtil · SD 1.5 LoRA, 30 stepsspeedup: 1.78 -> 2.46Improved
Otil instead of DistriFusionSD 1.5, 2 GPUs · COCO Captions 2014LPIPS: 0.041 -> 0.052Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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