Diffusion Model Inference Acceleration

Modern video diffusion models achieve higher generation quality through scaling, but this also increases inference cost. Although many acceleration methods have been proposed, a central challenge is that the most effective acceleration strategy is highly instance-specific: a recipe that works well for one combination of model, hardware, and inference configuration often does not transfer to another. Different models vary in architecture, numerical sensitivity, and attention concentration patterns. Inference settings differ in spatial and temporal resolution and video duration, while hardware platforms differ in memory hierarchy, supported numerical formats, and kernel throughput. These factors create a large tuning space, making manual performance engineering costly. We present Sol Video Inference Engine, an agentic, native, training-free acceleration framework for video diffusion models. It organizes five broadly applicable techniques, cache, sparse attention, token pruning, quantization, and kernel fusion, into an agentic acceleration stack for instance-specific optimization. For a concrete deployment target defined by a model, hardware platform, and serving configuration, parallel skill agents optimize the implementation of each technique, an agent integrator composes them into a global acceleration stack, and a human validator provides feedback on generation quality. We instantiate this workflow on three video models with different sizes and architectures: 64B Cosmos3-Super, 22B LTX-2.3, and 2B SANA-Video. With little human effort, the full stack achieves more than 2x end-to-end acceleration while maintaining near-lossless VBench quality, demonstrating the effectiveness of the agent framework for video diffusion acceleration.

Consistency distillation has significantly accelerated diffusion-model inference, but its sampling dynamics remain underexplored. We reveal an asymmetry: although Logit-Normal sampling priors work well for standard iterative generation, consistency distillation exhibits a different difficulty profile (e.g., U-shaped), with optimization bottlenecks concentrated at the boundary stages rather than intermediate steps. To address the limitations of static sampling under evolving learning demands, we propose Curvature-Adaptive Consistency Flow Matching (CACFM). By formulating distillation as a dynamic decision process, CACFM uses a lightweight reinforcement learning agent to probe Probability Flow ODE trajectories and construct an efficiency-oriented curriculum that prioritizes critical regions without manual scheduling. Combined with Flow-adapted DMD and adversarial consistency objectives, our RL-based scheduler achieves state-of-the-art results on large-scale models such as FLUX and SDXL, mitigating structural deformities and preserving high-frequency details in extreme few-step regimes.

Diffusion LLMs (dLLMs) improve GPU utilization over autoregressive decoding by generating multiple tokens per forward pass, but their KV cache still grows linearly with context, limiting throughput at long contexts. KV cache offloading to host DRAM alleviates this memory pressure, but the limited PCIe bandwidth necessitates recalling only a sparse subset of KV entries. In block dLLMs, the relevant KV entries remain consistent across denoising steps within a block, enabling high-accuracy selection by identifying the top-k entries once and reusing them throughout all denoising steps. This property appears attractive for offloading as it amortizes the selection overhead across the entire block, but it requires exact attention over the full KV cache, which is too expensive under offloading. We present HERALD, a KV offloading system for block dLLMs that resolves this through two opportunities that reduce the required selection compute by a factor of the block size and enable selection to be overlapped with denoising. Across three block dLLMs and five long context tasks, HERALD achieves near-lossless accuracy at 5-10% KV budget and up to 1.59x lower per block latency and 2.47x higher throughput over GPU-only inference, with speedups growing with context length.

We introduce UNITY, a Universal-to-Specialized adapter for efficient and scalable composite conditioning in diffusion based image generation. Unlike prior methods that train separate adapters for each conditioning modality, UNITY jointly learns shared semantics across multiple conditioning types and subsequently specializes without modifying the underlying architecture. The proposed two stage training paradigm consists of a Universal Stage that captures cross modal representations across all conditioning modalities using half of the total training steps, followed by a Specialization Stage that refines modality specific features using the remaining training budget. At the core of UNITY are the Morphable Attention Flow (MAF) Network and Morph Wrapper modules, which enable channel aware and spatially adaptive feature alignment through learnable flow fields and attention based fusion. This constant complexity formulation supports flexible operation under both single and composite conditioning settings while significantly reducing inference latency and memory consumption. Extensive experiments across multiple datasets demonstrate that UNITY achieves state of the art image fidelity while maintaining superior memory efficiency. Code: https://github.com/arya-domain/UNITY

While 10B-level industrial foundation models have pushed the boundaries of image inpainting, their prohibitive computational costs severely hinder practical deployment. Constructing a highly optimized task-specific specialist offers a promising solution; however, extreme structural compression inevitably triggers a severe representation bottleneck. To conquer this, we propose Moebius, a highly efficient lightweight inpainting framework. We systematically reconstruct the diffusion backbone by introducing the Local-$λ$ Mix Interaction ($LλMI$) block. Comprising Local-$λ$ and Interactive-$λ$ modules, it elegantly summarizes spatial contexts and global semantic priors into fixed-size linear matrices, preserving complex latent interactions while drastically shedding parameters. Furthermore, to unlock the full representational capacity of this highly compact architecture, we synergistically pair it with an adaptive multi-granularity distillation strategy. Operating strictly within the latent space to avoid expensive pixel-space decoding, this strategy dynamically balances multiple gradient-based losses to achieve high-fidelity alignment. Extensive experiments across natural and portrait benchmarks demonstrate that this optimal synergy enables Moebius to rival or even surpass the generation quality of the 10B-level industrial generalist FLUX.1-Fill-Dev. Remarkably, Moebius achieves this using less than 2% of the parameters (0.22B vs. 11.9B) while delivering a $>15\times$ acceleration in total inference time, setting a new efficiency standard for high-fidelity inpainting. Project page at https://hustvl.github.io/Moebius.

Video diffusion has quickly grown into a key generative serving workload, yet producing each clip demands many denoising iterations over large spatio-temporal latents, which puts low-latency inference out of reach on a single device. A denoising step is therefore typically distributed across multiple accelerators, and TPU sub-slices have become an attractive and practical fabric for doing so. Current auto-parallel systems, however, search almost exclusively over logical device meshes and disregard how a chosen sharding is actually laid out on the physical TPU interconnect -- an oversight that leaves large, topology-dependent performance on the table. We address this gap with AoiZora, a compiler-mediated topology planner built for low-latency video diffusion inference on TPU sub-slices. Its guiding principle is to reconnect logical sharding with physical placement by drawing on different points in the compilation flow: AoiZora first eliminates weak sharding candidates from inexpensive pre-compilation IRs, then compiles only the ones that survive and orders their physical placements using compiled HLO together with a topology-aware communication model. The winning plan is realized along the ordinary compiler path, leaving model code, compiler lowering, collective kernels, and network routing entirely intact. On TPU v5e sub-slices, AoiZora reduces Wan 2.1 one-step denoising latency by as much as 1.42x relative to existing solutions.

Diffusion large language models (dLLMs) offer a promising alternative to autoregressive decoding by iteratively refining masked sequences, enabling parallel token updates and bidirectional conditioning. Their practical efficiency, however, is limited by sampling procedures that execute a fixed number of reverse denoising steps selected before decoding, spending computation on already-stable positions and sometimes committing unstable ones too early. We present \textsc{LESS}, a training-free, model-agnostic adaptive sampler that treats token commitment as an online stopping problem. \textsc{LESS} implements mutual-stability sampling through a joint stability rule that makes a masked position eligible for unmasking only when its top-1 prediction has high confidence, its top-1 token persists across recent reverse steps, and its predictive distribution is stable under top-$K$ inter-step Jensen--Shannon divergence. We evaluate \textsc{LESS} on Dream-7B, LLaDA-8B, and LLaDA-1.5-8B, covering full-sequence diffusion and semi-autoregressive blockwise sampling regimes, across seven benchmarks spanning general knowledge, math, and code. \textsc{LESS} improves average accuracy over strong training-free adaptive samplers while using $72.1\%$ fewer reverse steps than fixed-budget decoding. Since each reverse step requires a Transformer forward pass, these step-count reductions translate into fewer forward evaluations, lower measured wall-clock latency, and lower estimated inference compute.

Diffusion Large Language Models (dLLMs) offer a promising avenue for parallel generation but face a trade-off between decoding speed and quality. While revocable decoding strategies attempt to mitigate errors by verifying and remasking tokens, they typically operate within a mixed-quality context. This leads to two critical failures: \textit{Error Propagation}, where new tokens absorb toxic information from erroneous context, and \textit{Local Error Reinforcement}, where errors mutually reinforce each other to evade detection. To alleviate these challenges, we propose ASRD (Anchor Supervised Revocable Decoding), a training-free framework that operates within the embedding space. ASRD explicitly decouples the decoding context into trusted \textit{Anchor Tokens}, which are identified via temporal consistency, and uncertain candidates. Leveraging a dynamic Anchor Tokens Cache, we introduce two complementary mechanisms: (1) Anchor-Guided Generation, which injects entropy-weighted anchor signals into masked positions to implicitly rectify attention toward the reliable global skeleton; and (2) Anchor-Perturbed Verification, which applies orthogonal perturbations to uncertain candidate tokens, destabilizing and remasking errors driven by fragile local consensus. Extensive experiments on math and coding benchmarks demonstrate that ASRD outperforms recent remasking baselines, achieving accuracy improvements of up to 6.4% while accelerating inference throughput by up to 7.2$\times$.The code is available at https://github.com/preordinary/ASRD.

Sparse attention accelerates Diffusion Transformers (DiTs) for video generation by computing only the important tokens while skipping the rest. The token selection strategy is key to balancing sparsity and accuracy. We formulate the token filtering process as a dual-goal optimization problem: maximizing sparsity and minimizing accuracy degradation. Existing algorithms cannot fulfill both objectives simultaneously. For example, Top-p only considers the accuracy constraint, while Top-k maintains a fixed computational budget but loosens the accuracy constraint. This paper demonstrates that maintaining a fixed recall rate is sufficient for ensuring accuracy, whereas a fixed threshold is suboptimal for reducing computational cost. Therefore, we propose a dynamic thresholding scheme to improve sparsity while maintaining the same level of accuracy. Furthermore, our algorithm is deeply integrated with Flash Attention (FA), eliminating the need for any additional masking computation overhead. Experimental results on Wan 2.2 validate that, compared to the BLASST algorithm which is also integrated with FA, our dynamic thresholding strategy enhances sparsity from 61.42% to 82% with a VBench metric drop of less than 5%. This results in an approximate 15% in attention computation and a $1.61\times$ increase in computational efficiency, which is 1.18x higher than that of BLASST.

Visual anagram is an intriguing form of art creation wherein a single image presents different conceptual interpretations under transformations such as flipping or rotation. Recent work has achieved visual anagram synthesis by leveraging pretrained text-to-image (T2I) diffusion models, yet still suffers from several key limitations including computational inefficiency, suboptimal aesthetic quality, and weak semantic fidelity and expressiveness. This work focuses on generating visual anagrams with substantially improved visual quality at minimal computational cost, thereby advancing intelligent creation of illusionary digital art. To increase image resolution while reducing time overhead, we adapt the cutting-edge parallel denoising algorithm from pixel-based T2I model to the adversarially distilled latent-based one, and accordingly propose a structure-semantic co-optimization (S2CO) framework to counteract the consequent visual degradation. As the core of our approach, S2CO framework comprises three key innovations: (\romannumeral1) null-text structure alignment optimization; (\romannumeral2) semantic enhancement optimization; (\romannumeral3) attention-guided noise fusion. Building upon these components, our method dubbed \textbf{S2CO-Anagram} is able to generate higher-resolution anagram images with noticeably superior visual harmony and semantic faithfulness than related SOTA approaches, all while achieving substantially faster inference speed. Code will be publicly available.

Diffusion models excel in Real-World Image Super-Resolution (Real-ISR) due to their powerful generative priors but suffer from slow iterative sampling. Although existing one-step distillation methods accelerate inference, they typically require auxiliary teacher models that inflate training memory and restrict scalability to large-scale architectures. Furthermore, these fixed-step models lack the flexibility to trade off speed for quality. In this paper, we propose TEASR, a training-efficient any-step diffusion framework for Real-ISR that enables both one-step and multi-step restoration within a unified model. Our key idea is to perform self-adversarial distillation within a single diffusion model, eliminating the need for auxiliary teachers or discriminators. Specifically, we propose a timestep-aware rectification strategy that stabilizes one-step generation across noise levels. These two designs further enables the distillation of 20B-parameter diffusion models on a single GPU, significantly improving training efficiency. Moreover, we introduce a dual-branch diffusion transformer with decoupled timestep condition to separate the current noise state and the denoising target to enhance sampling quality. Extensive experiments demonstrate that TEASR supports seamless any-step sampling and consistently outperforms state-of-the-art methods across multiple datasets.

Reconstructing dense 3D scenes from sparse LiDAR point clouds is a fundamental challenge in autonomous driving, where latent diffusion models offer a promising solution. However, existing approaches rely on object-level autoencoders that collapse into unstable global representations at outdoor scale and suffer from ground truth data corrupted by odometry drift that systematically degrades supervision quality. Furthermore, multi-step diffusion inference incurs prohibitive latency for real-time deployment. We propose a novel multi-token Gaussian VAE with cross-attention pooling for stable scene-scale LiDAR compression, combined with an anchor-based ICP ground truth refinement pipeline that eliminates drift-induced noise from training supervision. Together, these components enable a scaffold-free single-step diffusion completion model that achieves an approximately 16x reduction in squared Chamfer distance on SemanticKITTI seq. 08 (0.396 m^2 to 0.024 m^2), surpasses LiDiff and ScoreLiDAR by 17-19% and 10-11%, respectively, and operates at 25-143x lower inference latency. Our results demonstrate that data quality dominates model design in this regime and that multi-token latent spaces provide a stable first stage for latent diffusion-based scene completion.

Discrete diffusion language models enable parallel token generation, offering a pathway to low-latency decoding. However, selecting tokens independently by marginal confidence limits effective parallelism: tokens that appear reliable in isolation can form incompatible configurations when several positions are updated at once. We introduce a training-free decoding framework that coordinates these parallel updates. At each forward pass, the method assigns a commit score to each masked position and refines these scores using pairwise interactions derived from the model's predictive distributions. A variational relaxation yields a simple fixed-point update that suppresses conflicting simultaneous commitments within a single forward pass. This mechanism allows the decoder to commit more tokens in parallel while maintaining competitive generation quality. The method is lightweight, requires no auxiliary model or retraining, and drops into existing diffusion decoding pipelines without modification. Experiments on reasoning and code-generation benchmarks show consistent improvements in the quality-latency trade-off.

Diffusion Transformers with Mixture-of-Experts (DiT-MoE) improve model capacity under sparse activation, but diffusion inference is still bottlenecked by redundant computation across timesteps. Existing caching methods mainly operate at the token level, which becomes suboptimal in DiT-MoE because each token update is internally decomposed into multiple routed expert branches. Our analysis shows that cross-timestep redundancy in DiT-MoE is better characterized at the expert-branch level than at the whole-token level. Based on this observation, we propose MoECa, a fine-grained caching framework that performs branch-level feature reuse across timesteps. MoECa further introduces expert-aware adaptive control and synchronized cache updates across MoE and attention paths to maintain stable intermediate states. Experiments on multiple DiT-MoE models show that MoECa consistently achieves a better speed-quality trade-off than prior caching methods, with up to 2.83$\times$ inference speedup and minimal quality degradation.

Instruction-based image editing has made notable progress with recent advances in generative models. However, the quality of the edited result is still influenced by the randomly sampled initial noise, particularly in complex editing scenarios. An unsuitable initial noise may lead to unsatisfactory editing results. Recent inference-time scaling methods address this issue by sampling multiple initial noises and selecting better candidates. Nevertheless, most of them follow a decode-then-verify scheme which introduces an efficiency-accuracy trade-off. When decoding is performed after limited inference steps, the decoded images often remain too noisy for reliable assessment, whereas sufficiently denoised images require much higher computational cost. To address this issue, we propose VeriLatent, a plug-and-play adaptive inference-time scaling framework with early-step latent verification for image editing. Specifically, we propose a novel verifier that scores each initial noise through a latent-space editing activation map at an early stage. It identifies promising candidates by assessing whether they can induce an effective edit in the correct region. This enables efficient early pruning without decoding latents into images. Building on this, we further develop an adaptive search strategy for inference-time scaling. It allocates inference budgets according to editing difficulty, thereby reducing the number of function evaluations (NFE). Extensive experiments on multiple benchmarks and different base models demonstrate that VeriLatent consistently improves both editing performance and inference-time scaling efficiency.

Post-training INT8 (W8A8) quantization of diffusion transformers is widely deployed as a speed optimization, yet on consumer Ampere GPUs it is frequently slower than the FP8 and NF4 alternatives it is meant to beat. We trace this to a software artifact: the production "INT8" forward quantizes weights and activations only to immediately dequantize them back to bf16 and run a bf16 matrix multiply, never engaging the GPU's INT8 tensor cores, so the hardware's compute advantage is left entirely unrealized. We close this gap with a single fused Triton INT8 GEMM (int8xint8->int32 on Ampere tensor cores, with per-token x per-channel dequantization and bias folded into the epilogue, autotuned per GEMM shape) dropped into the Ideogram 4.0 diffusion transformer's linear layers in place of the dequantize-to-bf16 path. In the kernel, the int8xint8->int32 accumulation is bit-exact against torch._int_mm and the dequantized output matches the reference at cosine similarity 1.0 with no NaNs, running 2.8-4.2x faster than bf16 per GEMM. End to end it delivers a ~1.1x (~9-10%) speedup at 768px, and at 1024px it generates an image in 156.5 s on a single RTX 3090, faster than the single-card NF4 (164.5 s) and FP8 (172.9 s) baselines, at no measurable quality cost on these point estimates (PickScore/CLIPScore). INT8 thus goes from the slowest variant to the fastest, and 1024px becomes single-GPU feasible. The primary speed criterion (beat FP8, by ~9.5%) is comfortably met; the NF4 margin (~4.9%, single-run n=4) is within run-to-run variance we did not quantify and is best read as consistent with meeting the stretch target. We close with an honest deployment map: the win is specific to consumer Ampere, and on A100 and B200 the same kernel loses to those cards' fast native bf16/FP8 paths.

Creative image editing tools, such as Photoshop's Remove or Generative Fill buttons, are central to everyday customer use and account for a major share of traffic in Photoshop and Lightroom. However, current generative AI models face significant latency challenges, which become even more pronounced when transitioning from convolution-based U-Nets to Diffusion Transformers (DiTs). In our evaluation on hundreds of representative image editing samples spanning a wide range of mask ratios, the DiT module alone accounts for an average of 73% of the total model latency, even after being distilled from 50 timesteps down to 8 timesteps. To tackle this challenge, we propose $\textbf{HiLo-Token}$, an input-adaptive token compression framework that allocates more token budget to high-frequency, rich-context regions while assigning fewer tokens to low-frequency areas. Specifically, for the editing region specified by the user mask, we retain all tokens within a dilated mask to preserve strong locality and contextual relevance. Outside the editing region, we introduce a simple yet effective high-frequency token selection strategy based on spatial frequency to capture important local details, while using tokens from a 16x downsampled image to represent low-frequency components and preserve the blurry but global structure. Extensive experiments on production-level evaluation data validate the effectiveness of the proposed method, achieving 3.13x, 2.59x, and 1.67x DiT speedups on A100-80GB for image editing tasks across small, medium, and large mask ratio categories with average ratios of 6.38%, 15.92%, and 35.36%, respectively, without any regression in generation quality.

Diffusion Transformers (DiTs) have become the dominant architecture for image and video generation, creating growing demand for efficient DiT serving. Existing systems assign each request a fixed parallel configuration throughout its lifetime. However, DiT workloads exhibit substantial heterogeneity across requests, execution stages, and system conditions, making static parallelism inefficient and often leading to poor GPU utilization and degraded service quality. This paper argues that DiT serving should treat GPU parallelism as a first-class schedulable resource. We present GF-DiT, a policy-programmable runtime for elastic DiT serving that dynamically adapts the parallelism of running requests according to workload demands and service objectives. GF-DiT introduces an asynchronous execution abstraction that decomposes requests into independently schedulable trajectory tasks and enables online GPU reallocation. To make elastic parallelism practical, GF-DiT further proposes group-free collectives, a lightweight communication abstraction that supports low-overhead online formation and reconfiguration of arbitrary execution groups. We implement GF-DiT in vLLM-Omni and evaluate it on representative image and video diffusion workloads. Compared with fixed-pipeline execution with static parallelism, GF-DiT improves throughput by up to 6.01$\times$, reduces mean latency by up to 95%, lowers SLO violation rates by up to 90%, and reduces communication-group setup overhead from 778 ms to approximately 60 $μ$s. Our code is available at https://github.com/SJTU-Liquid/GF-DiT.

Step-level caching accelerates diffusion models by exploiting temporal redundancy across denoising steps. Existing methods make per-step cache decisions using threshold-based heuristics, without directly optimizing for final output quality. As a result, their inference latency varies across inputs and is difficult to control at deployment. In this work, we propose BudCache, which inverts this formulation: rather than letting per-step error thresholds dictate the runtime cost, we fix the compute budget in advance and search for the cache policy that best preserves the final output. To tackle the combinatorial complexity of step selection, we combine Simulated Annealing with deterministic Hill Climbing. This offline search identifies high-quality cache policies within minutes and introduces no online search or thresholding overhead during inference. When the compute budget is very tight, we further introduce cache-aware schedule alignment, which adapts the time discretization to the selected cache policy to reduce cache-induced trajectory mismatch. Experiments on FLUX.1-dev and Wan2.1 show that BudCache achieves better generation quality than heuristic caching baselines under the same inference budgets. Code is available at https://github.com/Westlake-AGI-Lab/BudCache

Speculative decoding speeds up LLM inference by using a draft model to generate tokens, with an acceptance-rejection scheme that ensures that the output matches the target distribution. Adapting this to continuous diffusions is difficult because speculative sampling requires drawing from a residual distribution. While straightforward in discrete spaces, efficiently sampling this residual in continuous space is non-trivial. Consequently, existing diffusion adaptations either use computationally inefficient sampling techniques or rely on an alternative scheme. In this work, we introduce a novel scheme that efficiently implements the original speculative sampling mechanism for diffusion models. Our approach offers a critical advantage over current methods: it enables us to adapt block verification from LLMs to diffusions -- which provably improves the acceptance rate of drafts. Furthermore, we formalize and analyze the Free Drafter, a heuristic self-speculative drafter for diffusions that requires no training. By enabling block verification, our Free Drafter yields up to a 6.3% speedup over existing speculative methods with no additional training and negligible overhead beyond the existing parallel verification pass.

Diffusion-based lip synchronization models achieve strong visual quality and audio-visual alignment, but full-sequence bidirectional attention and many denoising steps make them impractical for real-time inference. We present Lip Forcing, to our knowledge the first autoregressive diffusion method for video-to-video (V2V) lip synchronization, which distills a 14B audio-conditioned bidirectional video diffusion teacher into causal students. At inference, the students generate each chunk in only two denoising steps without inference-time CFG, enabling real-time lip synchronization. A lip-sync-specific teacher-trajectory analysis reveals a CFG fidelity-sync tradeoff: no-CFG predictions favor reference fidelity, whereas CFG-guided predictions favor synchronization within a mid-trajectory band. Lip Forcing translates this finding into three analysis-derived components: Sync-Window DMD, a two-step inference schedule, and a SyncNet-based reward. We validate Lip Forcing at two student scales, both distilled from the 14B teacher. The 1.3B student crosses into real-time streaming at 31 FPS, $17.6\times$ faster than its same-scale bidirectional model. The 14B student, the largest diffusion model reported for V2V lip synchronization, runs $39.8\times$ faster than its teacher at comparable reference fidelity. Time-to-first-frame is sub-millisecond at both scales, far below every diffusion baseline.

Masked diffusion language models can reduce inference steps by revealing multiple tokens per denoising iteration, but this parallelism is fragile: positions that are individually confident may be unsafe to commit together when their predictions are coupled. Existing training-free samplers such as Top-$k$, Fast-dLLM, and EB-Sampler mainly control how many tokens to reveal, while often ranking candidates by token-wise scores that ignore interactions within the selected set. We propose ADAS, a training-free reranking rule that leaves the base sampler's stopping rule unchanged and greedily discounts each token-wise confidence score according to its attention to already selected positions, weighted by their prediction uncertainty. Across LLaDA-8B-Base and Dream-7B-Base on the reasoning benchmarks GSM8K and MATH500 and the code benchmarks HumanEval and MBPP, plugging ADAS into all three samplers improves low-NFE performance at matched denoiser evaluations by $9.11$ and $10.46$ percentage points on average, respectively, with $3.1\%$ per-forward runtime overhead. Code is available at https://github.com/yusufsahin99/ADAS.

Diffusion large language models (dLLMs) re-encode the entire prefix at every denoising step, causing recomputation that scales quadratically with context length and becomes prohibitive for long-context scenarios. We propose Prefilling-dLLM, a training-free prefill-decode disaggregation framework for dLLMs that partitions the prefix into N chunks, caches their KV representations once, and selects the top-K most relevant chunks with intra-chunk token sparsity for decoding, showing that sparse prefilling can outperform dense attention while reducing per-step complexity from quadratic in the full sequence length to quadratic only in the decode length. On LongBench and InfiniteBench, Prefilling-dLLM achieves state-of-the-art quality among dLLM acceleration methods, and an attention kernel that parallelizes decoding over the non-contiguously cached chunk KV yields 9.1--28.0x speedup at 8K--32K contexts. We further show that beginning-of-sequence tokens prepended to each chunk act as periodic attention anchors that eliminate the lost-in-the-middle phenomenon. Code is available at https://github.com/menik1126/Prefilling-dLLM.

World-Action Models (WAMs) have emerged as a promising paradigm for embodied control by coupling future visual prediction with action generation. However, most existing WAMs rely on photorealistic future prediction, which incurs high inference latency and makes real-time robot deployment difficult. This motivates a more efficient WAM design that preserves the control benefits of future visual prediction while reducing its inference cost. We introduce Efficient-WAM, a World-Action Model that reduces the cost of future imagination while preserving its control benefit. Efficient-WAM improves inference efficiency via a compact video expert transferred from WAN-2.2-5B, token-sparse video latents, and asymmetric video-action denoising that allocates fewer sampling steps to video than to actions. Instead of optimizing the future branch for visual fidelity, Efficient-WAM treats future video prediction as a compact guidance signal for action generation. Comprehensive experiments on RoboTwin 2.0 and real-world manipulation tasks show that Efficient-WAM maintains strong action performance despite visibly coarse future predictions. While maintaining competitive control capabilities, our 1B-parameter model can reduce per-chunk latency to around 100 ms during physical deployment, achieving a 30x speedup over existing WAMs.

Real-time video restoration (VR) for live streams requires high-resolution outputs under strict per-frame latency constraints. Existing one-step diffusion-based VR models remain difficult to deploy on consumer-grade GPUs due to two main bottlenecks: quadratic spatial attention at high resolutions and the latency-memory overhead of large video autoencoders. We present SwiftVR, a streaming one-step generative VR framework that reduces both bottlenecks under a causal chunk-wise protocol. For attention, mask-free shifted-window self-attention gathers each spatial window into a dense tensor via deterministic indexing, keeping all attention calls on the dense scaled dot-product attention path without masks, cyclic shifts, padding, or hardware-specific sparse kernels. Because SwiftVR uses only standard dense SDPA calls, the trained model transfers to consumer GPUs without retraining or custom kernels. For autoencoding, a lightweight Restoration-aware Autoencoder enables fast chunk-wise decoding while preserving reconstruction quality. On a single H100, SwiftVR sustains 31FPS at 2560x1440 and 14FPS at 3840x2160, whereas all compared diffusion-based VR baselines exceed the memory limit at 4K. On a consumer RTX5090, SwiftVR reaches 26FPS at 1920x1080. To our knowledge, SwiftVR is the first generative VR model to achieve real-time 1080p streaming on a consumer-grade GPU, while attaining strong no-reference perceptual quality with lower inference cost. Project is available at https://h-oliday.github.io/SwiftVR.

While recent autoregressive video diffusion models achieve remarkable streaming quality, they remain confined to low resolutions (e.g., 480P), leaving efficient, scalable, real-time high-resolution video generation a fundamental open challenge. To bridge this gap, we present Ultra Flash, a cascaded streaming framework capable of real-time high-resolution video generation. Ultra Flash achieves ~30 FPS at 1K resolution and ~18 FPS at 2K resolution on a single GPU through three key contributions: (1) an architecture-preserving T2V-to-TV2V super-resolution training paradigm coupled with an AIGC-oriented data degradation pipeline that effectively preserves the generative capability of the base model, enabling enhanced high-resolution detail when cascaded after mainstream low-resolution generative models; (2) a causal streaming latent upsampler paired with a high-resolution decoder, which enhances spatiotemporal coherence while enabling efficient latent spatial scaling and precise high-resolution decoding with negligible computational overhead; and (3) a cascade high-resolution streaming video generation optimization scheme that first performs hybrid-reward-enhanced sparse causalization and single-step distillation of the super-resolution model, then introduces cascaded streaming self-forcing preference optimization with dynamic cache management, jointly enhancing overall coherence, improving quality, and enabling real-time high-resolution streaming video generation. Extensive experiments demonstrate that Ultra Flash reliably produces ultra-high-resolution streaming video while maintaining state-of-the-art visual quality and superior efficiency. Project Page: https://xin1u.github.io/UltraFlash/

Accurate quantification and uptake measurement in PET are critical for assessing disease progression and supporting clinical decision-making. While high-count PET provides reliable image quality, the associated radiation dose and prolonged acquisition remain significant clinical concerns, motivating the adoption of low-count protocols. Diffusion-model-based methods have demonstrated strong potential for restoring low-count PET to near high-count quality, but their iterative sampling procedure becomes prohibitively expensive when applied to high-resolution 3D PET volumes, introducing substantial inference latency that limits practical clinical deployment. To address these challenges, we propose a training-free Global-Local Skipping Strategy that accelerates diffusion model-based 3D PET denoising while simultaneously improving reconstruction quality. The proposed method is plug-and-play and directly applicable to pre-trained diffusion models without retraining or architectural modification. Specifically, we introduce: (i) a global denoising step skipping strategy that initializes the reverse diffusion process from an intermediate denoising step using a noise-consistent transformation of the low-count input, substantially reducing the number of required denoising steps; and (ii) a local feature reuse shortcut that reuses slowly-varying high-level U-Net features across neighboring denoising steps, further reducing per-step computation while preserving image fidelity. We evaluate the proposed approach on multiple PET tracers from in-house and public datasets, including 18F-FDG PET, 68Ga-DOTATATE PET, and 18F-PSMA PET, demonstrating consistent acceleration of over an order of magnitude alongside improved or comparable reconstruction performance relative to the full-step baseline. Blinded reader studies further confirm enhanced clinical confidence and perceived diagnostic quality.

Diffusion and flow generative models sample by integrating a learned ODE, but high quality still requires many sequential model evaluations. Solver learning reduces this cost by adapting scalar coefficients, timesteps, or both, while keeping the backbone model fixed. In this work, we identify a structural bottleneck in this update family: each step remains span-limited. Since the scalar-coefficient update lies in the span of buffered velocity evaluations, it can fit only the in-span component while leaving any out-of-span residual unreachable by scalar recombination alone. We propose SpanLift, a lightweight neural solver that augments scalar-coefficient updates with a spatial residual operator. SpanLift keeps a fixed base solver as an in-span prior and learns a spatial residual operator over the state and velocity buffer. The operator is trained by endpoint teacher matching, preserves the pretrained backbone, and adds no model NFEs. Empirically, the learned correction transfers across base solvers and is predominantly out-of-span. Across pixel-space diffusion, latent flow matching, and precipitation nowcasting, SpanLift achieves state-of-the-art few-step sampling. With only 3 NFE, it improves CIFAR-10 FID from 8.16 to 5.69 and ImageNet FID from 17.37 to 11.83.

Block-wise semi-autoregressive decoding is the standard inference paradigm for diffusion large language models (DLMs), but it imposes a strict dependency between blocks: the next block cannot begin until the current block is fully decoded or its denoising budget is exhausted. We observe that once a block exposes a reliable delimiter boundary or stable semantic prefix, continuation generation need not wait for every residual token to be resolved. We propose AsyncLane, a training-free decoding scheduler that decouples refinement from advancement. AsyncLane forks a generate lane at observed delimiter boundaries into a refine lane and a continuation generate lane: the prefix remains editable, while the continuation advances before prefix refinement finishes. The resulting lane tree records decoding dependencies and output order, while execution proceeds over the active lane set. To make this asynchronous schedule efficient under bidirectional attention, AsyncLane combines shared-prefix lane batching, lookahead draft reuse, cascading termination, and compact cache refresh with refresh-logit reuse, preventing model-call cost from scaling directly with the number of lanes. AsyncLane is a drop-in replacement for block-wise DLM samplers and requires no retraining. Experiments on mathematical reasoning and code generation show that AsyncLane consistently improves throughput while maintaining competitive quality. Across LLaDA and Dream backbones, AsyncLane achieves the highest TPS in all evaluated benchmark-length settings; relative to the fastest competing baseline, it reaches peak speedups of 2.95x on LLaDA and 3.04x on Dream, with especially large gains under longer generation budgets.

Diffusion language models (DLMs) offer substantial speed advantages through parallel decoding, but the lack of token dependencies limits generation quality compared to autoregressive (AR) models. Recent progress attempts to bridge the gap via importance sampling, with DLM being the proposal and AR being the target. However, due to the huge gap between their distributions, the sampling requires a large number of particles and is thus expensive to compute. In this paper, we introduce PoE-Bridge, a novel decoding framework that drastically improves generation speed and accuracy by introducing an intermediate distribution to bridge the gap. The distribution is constructed as a Product-of-Experts (PoE) of the DLM proposal and the AR target. With the intermediate distribution, we first use the DLM to draft multiple continuations in parallel, then apply rejection sampling to verify the drafted tokens and move the resulting candidates toward the PoE. We then use importance sampling to further correct the PoE-aligned candidates toward the AR target. We further propose several improved techniques, including mixed-temperature sampling for enhanced diversity and elastic rejection windows for reducing wasted verification. Empirically, PoE-Bridge achieves significantly improved accuracy with $5\times$ speedup over the standard DLM decoding approach, and recovers at least 95% of the target AR model's performance, efficiently advancing most of the quality gap on challenging mathematical reasoning and coding tasks. Our code is available at https://github.com/juntongshi48/poe-bridge.