Organizations: Cheriton School of Computer Science, University of Waterloo, Waterloo, ON, Canada · Department of Computer Science, George Mason University, Fairfax, VA, USA
Multi-round diffusion image editing can reduce computation by reusing intermediate activations, but retaining these activations across editing rounds introduces substantial memory and bandwidth overheads. Compression can reduce these overheads, but efficient cache construction is challenging: compressing only after editing requires buffering over 100 GB of activations per request. Moreover, the compressed cache must support selective reuse, as future masked edits require only a subset of the cached activations. It also exposes a scheduling challenge, as cached edits are faster than initial generations, causing naive policies to severely delay either initial generations or editing requests. In this paper, we aim to address both challenges through compression-informed serving. We develop TileCache, a serving system for multi-round diffusion editing. It compresses reusable activations and incorporates a scheduler that balances both edits and initial generation requests. Compared with the state-of-the-art image editing baseline, TileCache sustains over 2x the request arrival rate at 90% SLO attainment. It completes edits 2.4-4.5x and initial generations 1.3-2.2x faster than the baseline, while preserving image quality.
Figures & tables
Figure 1: Activation reuse reduces edit latency, but cache transfer introduces overhead. Latency comparison for FLUX.2 at 1024×1024 with 5.9% masked tokens.
Figure 2: Slowdowns due to FCFS and SJF scheduling.
Figure 3: Overview of TileCache in multi-round editing.
Figure 4: Serving performance on 4× H100 with 70% edits.
CLIP ↑
Deviation from the Diffusers edit (no reuse)
Error added by compression
PSNR ↑
SSIM ↑
FID ↓
TileCache vs. FlashPS edit
Model
Dataset
Ratio
Diffusers
FlashPS
TileCache
FlashPS
TileCache
FlashPS
TileCache
FlashPS
TileCache
PSNR ↑
SSIM ↑
FLUX.2 distilled
MagicBrush
4.6 ×
26.1
26.2
26.2
27.1
26.5
0.937
0.937
31.5
31.5
43.4
0.994
PIE-Bench
4.6 ×
26.2
26.0
26.0
21.6
21.6
0.868
0.868
60.6
60.8
38.2
0.989
ImgEdit (5 rounds)
4.6 ×
27.6
27.8
27.8
23.9
23.9
0.911
0.911
21.4
21.5
42.2
0.993
FLUX.2 base
MagicBrush
25.1 ×
26.0
26.1
26.1
33.4
33.1
0.974
0.974
9.8
9.6
46.4
0.996
Table 1: End-to-end edit quality of TileCache and FlashPS
Figure 5: Masked-area PSNR on MagicBrush.
Figure 6: Masked-area PSNR across 200 ImgEdit sessions.
Figure 7: Ablation study on TileCache components.
Figure 8: Scheduling comparison on 4× H100 with 70% edits.
Figure 9: GPU platform sensitivity with 70% edits.
Figure 10: Model sensitivity on H100 across diffusion models.
Figure 11: Edit-fraction and cache-size sensitivity on 4× H100.
Appendix figures & tables3 assets
Supplementary material from the paper’s appendix.
Appendix
0.25 requests/s
0.275 requests/s
Delay budget τ
Goodput (%)
SLO attainment (%)
Fairness F
SLO attainment (%)
Fairness F
1.5
-0.1
94.6
0.91
84.4
0.69
2 (default)
0
94.7
0.91
84.6
0.68
3
+0.4
95.3
0.86
84.8
0.68
5
+0.2
94.8
0.81
85.0
0.62
∞ (none)
+1.9
95.1
0.80
87.4
0.56
Appendix
Table A1: Delay-budget sensitivity on 4× H100.
Host cache (GB)
50
100
200
400
800
Hit rate (%)
32
58
82
93
96
Appendix
Table A2: Host-cache hit rate versus host-cache size at 0.25 requests/s on 4× H100 with 70% edits.
Figure A1: Masked-area PSNR on MagicBrush for distilled FLUX.2 at 4, 6, and 8 denoising steps.
Real-world image generation often involves multi-turn editing, where users iteratively modify small regions while most image content remains unchanged. However, existing diffusion transformer (DiT)-based editing pipelines recompute the entire image at every turn, causing substantial redundant computation. Existing DiT acceleration methods further ignore semantic correspondence across prompts, leading to unnecessary recomputation or unsafe reuse that harms editing quality. To address this, we propose RegionCache, a semantic-aware reuse framework for multi-turn image editing that selectively reuses diffusion states from unchanged regions. RegionCache detects reusable regions through semantic overlap between consecutive prompts and cross-attention localization, and adopts an adaptive reuse schedule based on prompt similarity and contextual consistency. Experiments on PixArt-alpha demonstrate that RegionCache achieves 1.43x--2.55x end-to-end speedup while maintaining comparable image quality. Code is available at https://github.com/hebutBryant/RegionCache.
Peizheng Li, Xin Ai, Hanyuan Liu +2
School of Computer Science and Engineering, Northeastern University, China · School of Computer Science
Diffusion transformers are widely used in image-editing services. To preserve both efficiency and privacy, existing systems commonly rely on cloud-device collaboration. Keeping request-specific inference on the device avoids sending personal images and prompts to the cloud. These workloads often use common templates, such as in face swapping and virtual try-on, enabling KV state to be precomputed and reused across requests. However, this reusable state is large and scales with the number of templates, making it impractical to store all of them on mobile devices. Storing them in the cloud addresses this capacity issue, but cache transfer is placed onto the inference path and can even be slower than on-device computation, especially under fluctuating mobile networks. We introduce SAKURA, a cloud-edge architecture that delivers reusable attention state to mobile devices under a bandwidth budget. It first uses a fine-grained token-importance-aware compression method to preserve image quality under limited transfer budgets. At runtime, SAKURA adaptively selects the cache representation for each upcoming shard based on network bandwidth. To enable efficient execution, we develop a pipelined runtime that overlaps cache transfer and reconstruction with on-device execution. We evaluate SAKURA using three editing tasks and two diffusion models, using a Snapdragon 8 Elite Gen 5 phone. SAKURA reduces latency by 1.82x and 3.27x over CacheGen adapted to DiT and on-device generation, respectively. It also reduces cache storage by 3.06x over uncompressed cache, while preserving image quality.
Feature caching accelerates Diffusion Transformers by reusing intermediate computations across denoising timesteps, yet existing methods share a fundamental blind spot: they treat the standard reference denoising trajectory as the optimisation target, approximating its per-step states and thus inheriting its output as an implicit quality ceiling. We argue that this target is arbitrary -- what truly matters is the ideal result, not the intermediate path taken to reach it. We propose IdealCache, a training-free framework that abandons step-wise imitation of the reference trajectory. Instead, it anchors caching on a per-input ensemble of ideal trajectories constructed from candidate target results, aggregates their geometry into a single graph, and discovers optimal non-uniform schedules via a constrained shortest path on this graph. By decoupling from the reference process, IdealCache breaks the performance bottleneck of trajectory-approximation; our schedules can attain near-lossless fidelity in generation and even surpass on task-specific metrics (e.g., maintaining higher semantic consistency in image editing). Without retraining, IdealCache achieves 6.35x speedup on Qwen-Image, 5.84x on Qwen-Image-Edit, and 4.32x on HunyuanVideo, with competitive or superior quality across tasks.
Shang Xue, Li Zhu, Ping Chen
Independent Researcher, Santa Clara, California, United States · Independent Researcher, Mountain View, California, United States · University of Electronic Science and Technology of China