Daily Papers

Vision transformers have emerged as a promising alternative to convolutional neural networks for various image analysis tasks, offering comparable or superior performance. However, one significant drawback of ViTs is their resource-intensive nature, leading to increased memory footprint, computation complexity, and power consumption. To democratize this high-performance technology and make it more environmentally friendly, it is essential to compress ViT models, reducing their resource requirements while maintaining high performance. In this paper, we introduce a new block-structured pruning to address the resource-intensive issue for ViTs, offering a balanced trade-off between accuracy and hardware acceleration. Unlike unstructured pruning or channel-wise structured pruning, block pruning leverages the block-wise structure of linear layers, resulting in more efficient matrix multiplications. To optimize this pruning scheme, our paper proposes a novel hardware-aware learning objective that simultaneously maximizes speedup and minimizes power consumption during inference, tailored to the block sparsity structure. This objective eliminates the need for empirical look-up tables and focuses solely on reducing parametrized layer connections. Moreover, our paper provides a lightweight algorithm to achieve post-training pruning for ViTs, utilizing second-order Taylor approximation and empirical optimization to solve the proposed hardware-aware objective. Extensive experiments on ImageNet are conducted across various ViT architectures, including DeiT-B and DeiT-S, demonstrating competitive performance with other pruning methods and achieving a remarkable balance between accuracy preservation and power savings. Especially, we achieve up to 3.93x and 1.79x speedups on dedicated hardware and GPUs respectively for DeiT-B, and also observe an inference power reduction by 1.4x on real-world GPUs.

Parameter-shared pre-trained language models (PLMs) have emerged as a successful approach in resource-constrained environments, enabling substantial reductions in model storage and memory costs without significant performance compromise. However, it is important to note that parameter sharing does not alleviate computational burdens associated with inference, thus impeding its practicality in situations characterized by limited stringent latency requirements or computational resources. Building upon neural ordinary differential equations (ODEs), we introduce a straightforward technique to enhance the inference efficiency of parameter-shared PLMs. Additionally, we propose a simple pre-training technique that leads to fully or partially shared models capable of achieving even greater inference acceleration. The experimental results demonstrate the effectiveness of our methods on both autoregressive and autoencoding PLMs, providing novel insights into more efficient utilization of parameter-shared models in resource-constrained settings.

As the training of giant dense models hits the boundary on the availability and capability of the hardware resources today, Mixture-of-Experts (MoE) models become one of the most promising model architectures due to their significant training cost reduction compared to a quality-equivalent dense model. Its training cost saving is demonstrated from encoder-decoder models (prior works) to a 5x saving for auto-aggressive language models (this work along with parallel explorations). However, due to the much larger model size and unique architecture, how to provide fast MoE model inference remains challenging and unsolved, limiting its practical usage. To tackle this, we present DeepSpeed-MoE, an end-to-end MoE training and inference solution as part of the DeepSpeed library, including novel MoE architecture designs and model compression techniques that reduce MoE model size by up to 3.7x, and a highly optimized inference system that provides 7.3x better latency and cost compared to existing MoE inference solutions. DeepSpeed-MoE offers an unprecedented scale and efficiency to serve massive MoE models with up to 4.5x faster and 9x cheaper inference compared to quality-equivalent dense models. We hope our innovations and systems help open a promising path to new directions in the large model landscape, a shift from dense to sparse MoE models, where training and deploying higher-quality models with fewer resources becomes more widely possible.

While Diffusion Transformers (DiTs) have achieved breakthroughs in video generation, this long sequence generation task remains constrained by the quadratic complexity of attention mechanisms, resulting in significant inference latency. Through detailed analysis of attention maps in Video Diffusion Transformer (vDiT), we identify three recurring sparsity patterns: diagonal, multi-diagonal, and vertical-stripe structures. And even 3-6\% attention heads can be skipped. Crucially, these patterns exhibit strong layer-depth and head-position correlations but show limited dependence on the input content. Leveraging these findings, we propose Sparse-vDiT, a sparsity acceleration framework for vDiT comprising: 1) Pattern-optimized sparse kernels that replace dense attention with computationally efficient implementations for each identified sparsity pattern. 2) An offline sparse diffusion search algorithm that selects the optimal sparse computation strategy per layer and head via hardware-aware cost modeling. After determining the optimal configuration, we fuse heads within the same layer that share the same attention strategy, enhancing inference efficiency. Integrated into state-of-the-art vDiT models (CogVideoX1.5, HunyuanVideo, and Wan2.1), Sparse-vDiT achieves 2.09times, 2.38times, and 1.67times theoretical FLOP reduction, and actual inference speedups of 1.76times, 1.85times, and 1.58times, respectively, while maintaining high visual fidelity, with PSNR values reaching 24.13, 27.09, and 22.59. Our work demonstrates that latent structural sparsity in vDiTs can be systematically exploited for long video synthesis.

Reinforcement learning agents based on Transformer architectures have achieved impressive performance on sequential decision-making tasks, but their reliance on dense matrix operations makes them ill-suited for energy-constrained, edge-oriented platforms. Spiking neural networks promise ultra-low-power, event-driven inference, yet no prior work has seamlessly merged spiking dynamics with return-conditioned sequence modeling. We present the Spiking Decision Transformer (SNN-DT), which embeds Leaky Integrate-and-Fire neurons into each self-attention block, trains end-to-end via surrogate gradients, and incorporates biologically inspired three-factor plasticity, phase-shifted spike-based positional encodings, and a lightweight dendritic routing module. Our implementation matches or exceeds standard Decision Transformer performance on classic control benchmarks (CartPole-v1, MountainCar-v0, Acrobot-v1, Pendulum-v1) while emitting fewer than ten spikes per decision, an energy proxy suggesting over four orders-of-magnitude reduction in per inference energy. By marrying sequence modeling with neuromorphic efficiency, SNN-DT opens a pathway toward real-time, low-power control on embedded and wearable devices.

The optimal training configurations of large language models (LLMs) with respect to model sizes and compute budgets have been extensively studied. But how to optimally configure LLMs during inference has not been explored in sufficient depth. We study compute-optimal inference: designing models and inference strategies that optimally trade off additional inference-time compute for improved performance. As a first step towards understanding and designing compute-optimal inference methods, we assessed the effectiveness and computational efficiency of multiple inference strategies such as Greedy Search, Majority Voting, Best-of-N, Weighted Voting, and their variants on two different Tree Search algorithms, involving different model sizes and computational budgets. We found that a smaller language model with a novel tree search algorithm typically achieves a Pareto-optimal trade-off. These results highlight the potential benefits of deploying smaller models equipped with more sophisticated decoding algorithms in budget-constrained scenarios, e.g., on end-devices, to enhance problem-solving accuracy. For instance, we show that the Llemma-7B model can achieve competitive accuracy to a Llemma-34B model on MATH500 while using 2times less FLOPs. Our findings could potentially apply to any generation task with a well-defined measure of success.

Energy efficiency of Convolutional Neural Networks (CNNs) has become an important area of research, with various strategies being developed to minimize the power consumption of these models. Previous efforts, including techniques like model pruning, quantization, and hardware optimization, have made significant strides in this direction. However, there remains a need for more effective on device AI solutions that balance energy efficiency with model performance. In this paper, we propose a novel approach to reduce the energy requirements of inference of CNNs. Our methodology employs two small Complementary CNNs that collaborate with each other by covering each other's "weaknesses" in predictions. If the confidence for a prediction of the first CNN is considered low, the second CNN is invoked with the aim of producing a higher confidence prediction. This dual-CNN setup significantly reduces energy consumption compared to using a single large deep CNN. Additionally, we propose a memory component that retains previous classifications for identical inputs, bypassing the need to re-invoke the CNNs for the same input, further saving energy. Our experiments on a Jetson Nano computer demonstrate an energy reduction of up to 85.8% achieved on modified datasets where each sample was duplicated once. These findings indicate that leveraging a complementary CNN pair along with a memory component effectively reduces inference energy while maintaining high accuracy.

In recent years, large language models have demonstrated remarkable performance across various natural language processing (NLP) tasks. However, deploying these models for real-world applications often requires efficient inference solutions to handle the computational demands. In this paper, we explore the utilization of CPUs for accelerating the inference of large language models. Specifically, we introduce a parallelized approach to enhance throughput by 1) Exploiting the parallel processing capabilities of modern CPU architectures, 2) Batching the inference request. Our evaluation shows the accelerated inference engine gives an 18-22x improvement in the generated token per sec. The improvement is more with longer sequence and larger models. In addition to this, we can also run multiple workers in the same machine with NUMA node isolation to further improvement in tokens/s. Table 2, we have received 4x additional improvement with 4 workers. This would also make Gen-AI based products and companies environment friendly, our estimates shows that CPU usage for Inference could reduce the power consumption of LLMs by 48.9% while providing production ready throughput and latency.

Reducing inference time and energy usage while maintaining prediction accuracy has become a significant concern for deep neural networks (DNN) inference on resource-constrained edge devices. To address this problem, we propose a novel approach based on "converting" autoencoder and lightweight DNNs. This improves upon recent work such as early-exiting framework and DNN partitioning. Early-exiting frameworks spend different amounts of computation power for different input data depending upon their complexity. However, they can be inefficient in real-world scenarios that deal with many hard image samples. On the other hand, DNN partitioning algorithms that utilize the computation power of both the cloud and edge devices can be affected by network delays and intermittent connections between the cloud and the edge. We present CBNet, a low-latency and energy-efficient DNN inference framework tailored for edge devices. It utilizes a "converting" autoencoder to efficiently transform hard images into easy ones, which are subsequently processed by a lightweight DNN for inference. To the best of our knowledge, such autoencoder has not been proposed earlier. Our experimental results using three popular image-classification datasets on a Raspberry Pi 4, a Google Cloud instance, and an instance with Nvidia Tesla K80 GPU show that CBNet achieves up to 4.8x speedup in inference latency and 79% reduction in energy usage compared to competing techniques while maintaining similar or higher accuracy.

This paper introduces PowerInfer-2, a framework designed for high-speed inference of Large Language Models (LLMs) on smartphones, particularly effective for models whose sizes exceed the device's memory capacity. The key insight of PowerInfer-2 is to utilize the heterogeneous computation, memory, and I/O resources in smartphones by decomposing traditional matrix computations into fine-grained neuron cluster computations. Specifically, PowerInfer-2 features a polymorphic neuron engine that adapts computational strategies for various stages of LLM inference. Additionally, it introduces segmented neuron caching and fine-grained neuron-cluster-level pipelining, which effectively minimize and conceal the overhead caused by I/O operations. The implementation and evaluation of PowerInfer-2 demonstrate its capability to support a wide array of LLM models on two smartphones, achieving up to a 29.2x speed increase compared with state-of-the-art frameworks. Notably, PowerInfer-2 is the first system to serve the TurboSparse-Mixtral-47B model with a generation rate of 11.68 tokens per second on a smartphone. For models that fit entirely within the memory, PowerInfer-2 can achieve approximately a 40% reduction in memory usage while maintaining inference speeds comparable to llama.cpp and MLC-LLM. For more details, including a demonstration video, please visit the project site at www.powerinfer.ai/v2.

Vision Language Models (VLMs) have demonstrated strong capabilities across various visual understanding and reasoning tasks. However, their real-world deployment is often constrained by high latency during inference due to substantial compute required to process the large number of input tokens (predominantly from the image) by the LLM. To reduce inference costs, one can either downsize the LLM or reduce the number of input image-tokens, the latter of which has been the focus of many recent works around token compression. However, it is unclear what the optimal trade-off is, as both the factors directly affect the VLM performance. We first characterize this optimal trade-off between the number of visual tokens and LLM parameters by establishing scaling laws that capture variations in performance with these two factors. Our results reveal a surprising trend: for visual reasoning tasks, the inference-optimal behavior in VLMs, i.e., minimum downstream error at any given fixed inference compute, is achieved when using the largest LLM that fits within the inference budget while minimizing visual token count - often to a single token. While the token reduction literature has mainly focused on maintaining base model performance by modestly reducing the token count (e.g., 5-10times), our results indicate that the compute-optimal inference regime requires operating under even higher token compression ratios. Based on these insights, we take some initial steps towards building approaches tailored for high token compression settings. Code is available at https://github.com/locuslab/llava-token-compression.

As neural networks grow in size and complexity, inference speeds decline. To combat this, one of the most effective compression techniques -- channel pruning -- removes channels from weights. However, for multi-branch segments of a model, channel removal can introduce inference-time memory copies. In turn, these copies increase inference latency -- so much so that the pruned model can be slower than the unpruned model. As a workaround, pruners conventionally constrain certain channels to be pruned together. This fully eliminates memory copies but, as we show, significantly impairs accuracy. We now have a dilemma: Remove constraints but increase latency, or add constraints and impair accuracy. In response, our insight is to reorder channels at export time, (1) reducing latency by reducing memory copies and (2) improving accuracy by removing constraints. Using this insight, we design a generic algorithm UPSCALE to prune models with any pruning pattern. By removing constraints from existing pruners, we improve ImageNet accuracy for post-training pruned models by 2.1 points on average -- benefiting DenseNet (+16.9), EfficientNetV2 (+7.9), and ResNet (+6.2). Furthermore, by reordering channels, UPSCALE improves inference speeds by up to 2x over a baseline export.

Large language models (LLMs) are widely applied in chatbots, code generators, and search engines. Workloads such as chain-of-thought, complex reasoning, and agent services significantly increase the inference cost by invoking the model repeatedly. Optimization methods such as parallelism, compression, and caching have been adopted to reduce costs, but the diverse service requirements make it hard to select the right method. Recently, specialized LLM inference engines have emerged as a key component for integrating the optimization methods into service-oriented infrastructures. However, a systematic study on inference engines is still lacking. This paper provides a comprehensive evaluation of 25 open-source and commercial inference engines. We examine each inference engine in terms of ease-of-use, ease-of-deployment, general-purpose support, scalability, and suitability for throughput- and latency-aware computation. Furthermore, we explore the design goals of each inference engine by investigating the optimization techniques it supports. In addition, we assess the ecosystem maturity of open source inference engines and handle the performance and cost policy of commercial solutions. We outline future research directions that include support for complex LLM-based services, support of various hardware, and enhanced security, offering practical guidance to researchers and developers in selecting and designing optimized LLM inference engines. We also provide a public repository to continually track developments in this fast-evolving field: https://github.com/sihyeong/Awesome-LLM-Inference-Engine

Specializing large language models (LLMs) for local deployment in domain-specific use cases is necessary for strong performance while meeting latency and privacy constraints. However, conventional task-specific adaptation approaches do not show simultaneous memory saving and inference speedup at deployment time. Practical compression techniques like quantization and pruning require dedicated hardware or kernel support to achieve measured inference speedup. We develop TrimLLM based on the layer-wise specialization phenomenon we empirically observed and verified on contemporary LLMs. TrimLLM reduces the depth of LLMs via progressive layer dropping. We show it retains LLMs' capacity in specific domains and achieves inference speedup irrespective of hardware and deep learning frameworks. We evaluated TrimLLM on LLMs of various sizes for inference; models adapted on medical, legal, and financial datasets all demonstrate 2.1-5.7times inference speedup on consumer GPUs and up to 3.1times speedup on A100 when compared to state-of-the-art model compression algorithms, with no loss in accuracy at 50sim60\% model compression ratio.

Inference-time scaling trades efficiency for increased reasoning accuracy by generating longer or more parallel sequences. However, in Transformer LLMs, generation cost is bottlenecked by the size of the key-value (KV) cache, rather than the number of generated tokens. Hence, we explore inference-time hyper-scaling: by compressing the KV cache, we can generate more tokens within the same compute budget and further improve the accuracy of scaled inference. The success of this approach, however, hinges on the ability of compression methods to preserve accuracy even at high compression ratios. To make hyper-scaling practical, we introduce Dynamic Memory Sparsification (DMS), a novel method for sparsifying KV caches that only requires 1K training steps to achieve 8times compression, while maintaining better accuracy than training-free sparse attention. Instead of prematurely discarding cached tokens, DMS delays token eviction, implicitly merging representations and preserving critical information. We demonstrate the effectiveness of inference-time hyper-scaling with DMS on multiple families of LLMs, showing that it boosts accuracy for comparable inference runtime and memory load. For instance, we enhance Qwen-R1 32B by an average of 9.1 points on AIME 24, 7.6 on GPQA, and 9.6 on LiveCodeBench across compute budgets.

Large Reasoning Models (LRMs) achieve remarkable performance by explicitly generating multi-step chains of thought, but this capability incurs substantial inference latency and computational cost. Collaborative inference offers a promising solution by selectively allocating work between lightweight and large models, yet a fundamental challenge remains: determining when a reasoning step requires the capacity of a large model or the efficiency of a small model. Existing routing strategies either rely on local token probabilities or post-hoc verification, introducing significant inference overhead. In this work, we propose a novel perspective on step-wise collaboration: the difficulty of a reasoning step can be inferred from its very first token. Inspired by the "Aha Moment" phenomenon in LRMs, we show that the entropy of the initial token serves as a strong predictor of step difficulty. Building on this insight, we introduce GlimpRouter, a training-free step-wise collaboration framework. GlimpRouter employs a lightweight model to generate only the first token of each reasoning step and routes the step to a larger model only when the initial token entropy exceeds a threshold. Experiments on multiple benchmarks demonstrate that our approach significantly reduces inference latency while preserving accuracy. For instance, GlimpRouter attains a substantial 10.7% improvement in accuracy while reducing inference latency by 25.9% compared to a standalone large model on AIME25. These results suggest a simple yet effective mechanism for reasoning: allocating computation based on a glimpse of thought rather than full-step evaluation.

The emergence of large-scale Mixture of Experts (MoE) models has marked a significant advancement in artificial intelligence, offering enhanced model capacity and computational efficiency through conditional computation. However, the deployment and inference of these models present substantial challenges in terms of computational resources, latency, and energy efficiency. This comprehensive survey systematically analyzes the current landscape of inference optimization techniques for MoE models across the entire system stack. We first establish a taxonomical framework that categorizes optimization approaches into model-level, system-level, and hardware-level optimizations. At the model level, we examine architectural innovations including efficient expert design, attention mechanisms, various compression techniques such as pruning, quantization, and knowledge distillation, as well as algorithm improvement including dynamic routing strategies and expert merging methods. At the system level, we investigate distributed computing approaches, load balancing mechanisms, and efficient scheduling algorithms that enable scalable deployment. Furthermore, we delve into hardware-specific optimizations and co-design strategies that maximize throughput and energy efficiency. This survey not only provides a structured overview of existing solutions but also identifies key challenges and promising research directions in MoE inference optimization. Our comprehensive analysis serves as a valuable resource for researchers and practitioners working on large-scale deployment of MoE models in resource-constrained environments. To facilitate ongoing updates and the sharing of cutting-edge advances in MoE inference optimization research, we have established a repository accessible at https://github.com/MoE-Inf/awesome-moe-inference/.

Scaling model size and training data has led to great advances in the performance of Large Language Models (LLMs). However, the diminishing returns of this approach necessitate alternative methods to improve model capabilities, particularly in tasks requiring advanced reasoning. Large reasoning models, which leverage long chain-of-thoughts, bring unprecedented breakthroughs in problem-solving capabilities but at a substantial deployment cost associated to longer generations. Reducing inference costs is crucial for the economic feasibility, user experience, and environmental sustainability of these models. In this work, we propose to train large reasoning models to reason efficiently. More precisely, we use reinforcement learning (RL) to train reasoning models to dynamically allocate inference-time compute based on task complexity. Our method incentivizes models to minimize unnecessary computational overhead while maintaining accuracy, thereby achieving substantial efficiency gains. It enables the derivation of a family of reasoning models with varying efficiency levels, controlled via a single hyperparameter. Experiments on two open-weight large reasoning models demonstrate significant reductions in inference cost while preserving most of the accuracy.

Large language models (LLMs) have shown exceptional performance and vast potential across diverse tasks. However, the deployment of LLMs with high performance in low-resource environments has garnered significant attention in the industry. When GPU hardware resources are limited, we can explore alternative options on CPUs. To mitigate the financial burden and alleviate constraints imposed by hardware resources, optimizing inference performance is necessary. In this paper, we introduce an easily deployable inference performance optimization solution aimed at accelerating LLMs on CPUs. In this solution, we implement an effective way to reduce the KV cache size while ensuring precision. We propose a distributed inference optimization approach and implement it based on oneAPI Collective Communications Library. Furthermore, we propose optimization approaches for LLMs on CPU, and conduct tailored optimizations for the most commonly used models. The code is open-sourced at https://github.com/intel/xFasterTransformer.

The recent surge of open-source large language models (LLMs) enables developers to create AI-based solutions while maintaining control over aspects such as privacy and compliance, thereby providing governance and ownership of the model deployment process. To utilize these LLMs, inference engines are needed. These engines load the model's weights onto available resources, such as GPUs, and process queries to generate responses. The speed of inference, or performance, of the LLM, is critical for real-time applications, as it computes millions or billions of floating point operations per inference. Recently, advanced inference engines such as vLLM have emerged, incorporating novel mechanisms such as efficient memory management to achieve state-of-the-art performance. In this paper, we analyze the performance, particularly the throughput (tokens generated per unit of time), of 20 LLMs using two inference libraries: vLLM and HuggingFace's pipelines. We investigate how various hyperparameters, which developers must configure, influence inference performance. Our results reveal that throughput landscapes are irregular, with distinct peaks, highlighting the importance of hyperparameter optimization to achieve maximum performance. We also show that applying hyperparameter optimization when upgrading or downgrading the GPU model used for inference can improve throughput from HuggingFace pipelines by an average of 9.16% and 13.7%, respectively.

Inference-time optimization scales computation to derive deliberate reasoning steps for effective performance. While previous search-based strategies address the short-sightedness of auto-regressive generation, the vast search space leads to excessive exploration and insufficient exploitation. To strike an efficient balance to derive the optimal step, we frame the decoding strategy as foresight sampling, leveraging simulated future steps to obtain globally optimal step estimation. Built on it, we propose a novel decoding strategy, named phi-Decoding. To provide a precise and expressive estimation of step value, phi-Decoding approximates two distributions via foresight and clustering. Sampling from the joint distribution, the optimal steps can be selected for exploitation. To support adaptive computation allocation, we propose in-width and in-depth pruning strategies, featuring a light-weight solution to achieve inference efficiency. Extensive experiments across seven benchmarks show phi-Decoding outperforms strong baselines in both performance and efficiency. Additional analysis demonstrates its generalization across various LLMs and scalability across a wide range of computing budgets. The code will be released at https://github.com/xufangzhi/phi-Decoding, and the open-source PyPI package is coming soon.

Recent innovation in large language models (LLMs), and their myriad use-cases have rapidly driven up the compute capacity demand for datacenter GPUs. Several cloud providers and other enterprises have made substantial plans of growth in their datacenters to support these new workloads. One of the key bottleneck resources in datacenters is power, and given the increasing model sizes of LLMs, they are becoming increasingly power intensive. In this paper, we show that there is a significant opportunity to oversubscribe power in LLM clusters. Power oversubscription improves the power efficiency of these datacenters, allowing more deployable servers per datacenter, and reduces the deployment time, since building new datacenters is slow. We extensively characterize the power consumption patterns of a variety of LLMs and their configurations. We identify the differences between the inference and training power consumption patterns. Based on our analysis of these LLMs, we claim that the average and peak power utilization in LLM clusters for inference should not be very high. Our deductions align with the data from production LLM clusters, revealing that inference workloads offer substantial headroom for power oversubscription. However, the stringent set of telemetry and controls that GPUs offer in a virtualized environment, makes it challenging to have a reliable and robust power oversubscription mechanism. We propose POLCA, our framework for power oversubscription that is robust, reliable, and readily deployable for GPU clusters. Using open-source models to replicate the power patterns observed in production, we simulate POLCA and demonstrate that we can deploy 30% more servers in the same GPU cluster for inference, with minimal performance loss

This paper introduces PowerInfer, a high-speed Large Language Model (LLM) inference engine on a personal computer (PC) equipped with a single consumer-grade GPU. The key underlying the design of PowerInfer is exploiting the high locality inherent in LLM inference, characterized by a power-law distribution in neuron activation. This distribution indicates that a small subset of neurons, termed hot neurons, are consistently activated across inputs, while the majority, cold neurons, vary based on specific inputs. PowerInfer exploits such an insight to design a GPU-CPU hybrid inference engine: hot-activated neurons are preloaded onto the GPU for fast access, while cold-activated neurons are computed on the CPU, thus significantly reducing GPU memory demands and CPU-GPU data transfers. PowerInfer further integrates adaptive predictors and neuron-aware sparse operators, optimizing the efficiency of neuron activation and computational sparsity. Evaluation shows that PowerInfer attains an average token generation rate of 13.20 tokens/s, with a peak of 29.08 tokens/s, across various LLMs (including OPT-175B) on a single NVIDIA RTX 4090 GPU, only 18% lower than that achieved by a top-tier server-grade A100 GPU. This significantly outperforms llama.cpp by up to 11.69x while retaining model accuracy.

The subject of green AI has been gaining attention within the deep learning community given the recent trend of ever larger and more complex neural network models. Existing solutions for reducing the computational load of training at inference time usually involve pruning the network parameters. Pruning schemes often create extra overhead either by iterative training and fine-tuning for static pruning or repeated computation of a dynamic pruning graph. We propose a new parameter pruning strategy for learning a lighter-weight sub-network that minimizes the energy cost while maintaining comparable performance to the fully parameterised network on given downstream tasks. Our proposed pruning scheme is green-oriented, as it only requires a one-off training to discover the optimal static sub-networks by dynamic pruning methods. The pruning scheme consists of a binary gating module and a novel loss function to uncover sub-networks with user-defined sparsity. Our method enables pruning and training simultaneously, which saves energy in both the training and inference phases and avoids extra computational overhead from gating modules at inference time. Our results on CIFAR-10 and CIFAR-100 suggest that our scheme can remove 50% of connections in deep networks with less than 1% reduction in classification accuracy. Compared to other related pruning methods, our method demonstrates a lower drop in accuracy for equivalent reductions in computational cost.

The performance of neural networks improves when more parameters are used. However, the model sizes are constrained by the available on-device memory during training and inference. Although applying techniques like quantization can alleviate the constraint, they suffer from performance degradation. In this work, we introduce NeuZip, a new weight compression scheme based on the entropy of floating-point numbers in neural networks. With NeuZip, we are able to achieve memory-efficient training and inference without sacrificing performance. Notably, we significantly reduce the memory footprint of training a Llama-3 8B model from 31GB to less than 16GB, while keeping the training dynamics fully unchanged. In inference, our method can reduce memory usage by more than half while maintaining near-lossless performance. Our code is publicly available.

Membership inference attacks aim to detect if a particular data point was used in training a model. We design a novel statistical test to perform robust membership inference attacks (RMIA) with low computational overhead. We achieve this by a fine-grained modeling of the null hypothesis in our likelihood ratio tests, and effectively leveraging both reference models and reference population data samples. RMIA has superior test power compared with prior methods, throughout the TPR-FPR curve (even at extremely low FPR, as low as 0). Under computational constraints, where only a limited number of pre-trained reference models (as few as 1) are available, and also when we vary other elements of the attack (e.g., data distribution), our method performs exceptionally well, unlike prior attacks that approach random guessing. RMIA lays the groundwork for practical yet accurate data privacy risk assessment in machine learning.

Inference-time scaling has emerged as a powerful alternative to parameter scaling for improving language model performance on complex reasoning tasks. While existing methods have shown strong performance gains under fixed compute budgets, there has been little focus on optimally utilizing that budget during inference. In this work, we introduce A*-decoding, a search-based inference-time strategy that builds on the A* search algorithm to optimally utilize a fixed compute budget by prioritizing high-quality reasoning paths during generation. We frame language model decoding as a structured search in a state space of partial solutions, applying the A* transition model to identify promising continuations guided by an external process supervision signal. In our experiments, A*-decoding reaches the performance levels of strong inference scaling baselines like best-of-N and particle filtering while using up to 3x fewer tokens and 30% fewer PRM passes under equivalent compute budgets. On the MATH500 and AIME 2024 benchmarks, A*-decoding enables Llama-3.2-1B-Instruct to match the performance of the 70x larger Llama-3.1-70B-Instruct, and allows Qwen3-1.7B to reach o1-like reasoning accuracy. These results highlight the power of structured search in decoding, offering an alternative to brute-force sampling or scale-driven gains. Our work demonstrates how thoughtful inference-time strategies can enhance reasoning in SLMs, pointing toward future advances in more efficient and scalable language model deployment.

Large Language Models (LLMs) have achieved great success in solving difficult tasks across many domains, but such success comes with a high computation cost, and inference latency. As developers and third parties customize these models, the need to provide efficient inference has increased. Many efforts have attempted to reduce inference cost through model compression techniques such as pruning and distillation. However, these techniques either require labeled data, or are time-consuming as they require the compressed model to be retrained to regain accuracy. In this paper, we propose a gradient-free structured pruning framework that uses only unlabeled data. An evaluation on the GLUE and SQuAD benchmarks using BERT_{BASE} and DistilBERT illustrates the effectiveness of the proposed approach. By only using the weights of the pre-trained model and unlabeled data, in a matter of a few minutes on a single GPU, up to 40% of the original FLOP count can be reduced with less than a 4% accuracy loss across all tasks considered.

This work focuses on reducing neural network size, which is a major driver of neural network execution time, power consumption, bandwidth, and memory footprint. A key challenge is to reduce size in a manner that can be exploited readily for efficient training and inference without the need for specialized hardware. We propose Self-Compression: a simple, general method that simultaneously achieves two goals: (1) removing redundant weights, and (2) reducing the number of bits required to represent the remaining weights. This is achieved using a generalized loss function to minimize overall network size. In our experiments we demonstrate floating point accuracy with as few as 3% of the bits and 18% of the weights remaining in the network.

With the advancement of large language models (LLMs), solving complex reasoning tasks has gained increasing attention. Inference-time computation methods (e.g., Best-of-N, beam search, et al.) are particularly valuable as they can enhance reasoning performance without modifying model parameters or requiring additional training. However, these techniques come with implementation challenges, and most existing methods remain at the proof-of-concept stage with limited practical adoption due to their computational complexity and varying effectiveness across different tasks. In this paper, we investigate and benchmark diverse inference-time computation strategies across reasoning tasks of varying complexity. Since most current methods rely on a proposer-verifier pipeline that first generates candidate solutions (e.g., reasoning solutions) and then selects the best one based on reward signals (e.g., RLHF rewards, process rewards), our research focuses on optimizing both candidate solution generation (e.g., instructing prompts, hyperparameters such as temperature and top-p) and reward mechanisms (e.g., self-evaluation, reward types). Through extensive experiments (more than 20,000 A100-80G GPU hours with over 1,000 experiments) across a variety of models (e.g., Llama, Qwen, and Mistral families) of various sizes, our ablation studies reveal that previously overlooked strategies can significantly enhance performance (e.g., tuning temperature can improve reasoning task performance by up to 5%). Furthermore, we establish a standardized benchmark for inference-time computation by systematically evaluating six representative methods across eight reasoning tasks. These findings provide a stronger foundation for future research. The code is available at https://github.com/usail-hkust/benchmark_inference_time_computation_LLM

Recently, Zaremba et al. demonstrated that increasing inference-time computation improves robustness in large proprietary reasoning LLMs. In this paper, we first show that smaller-scale, open-source models (e.g., DeepSeek R1, Qwen3, Phi-reasoning) can also benefit from inference-time scaling using a simple budget forcing strategy. More importantly, we reveal and critically examine an implicit assumption in prior work: intermediate reasoning steps are hidden from adversaries. By relaxing this assumption, we identify an important security risk, intuitively motivated and empirically verified as an inverse scaling law: if intermediate reasoning steps become explicitly accessible, increased inference-time computation consistently reduces model robustness. Finally, we discuss practical scenarios where models with hidden reasoning chains are still vulnerable to attacks, such as models with tool-integrated reasoning and advanced reasoning extraction attacks. Our findings collectively demonstrate that the robustness benefits of inference-time scaling depend heavily on the adversarial setting and deployment context. We urge practitioners to carefully weigh these subtle trade-offs before applying inference-time scaling in security-sensitive, real-world applications.

We examine the reasoning and planning capabilities of large language models (LLMs) in solving complex tasks. Recent advances in inference-time techniques demonstrate the potential to enhance LLM reasoning without additional training by exploring intermediate steps during inference. Notably, OpenAI's o1 model shows promising performance through its novel use of multi-step reasoning and verification. Here, we explore how scaling inference-time techniques can improve reasoning and planning, focusing on understanding the tradeoff between computational cost and performance. To this end, we construct a comprehensive benchmark, known as Sys2Bench, and perform extensive experiments evaluating existing inference-time techniques on eleven diverse tasks across five categories, including arithmetic reasoning, logical reasoning, common sense reasoning, algorithmic reasoning, and planning. Our findings indicate that simply scaling inference-time computation has limitations, as no single inference-time technique consistently performs well across all reasoning and planning tasks.

Scaling the amount of compute used to train language models has dramatically improved their capabilities. However, when it comes to inference, we often limit the amount of compute to only one attempt per problem. Here, we explore inference compute as another axis for scaling by increasing the number of generated samples. Across multiple tasks and models, we observe that coverage - the fraction of problems solved by any attempt - scales with the number of samples over four orders of magnitude. In domains like coding and formal proofs, where all answers can be automatically verified, these increases in coverage directly translate into improved performance. When we apply repeated sampling to SWE-bench Lite, the fraction of issues solved with DeepSeek-V2-Coder-Instruct increases from 15.9% with one sample to 56% with 250 samples, outperforming the single-attempt state-of-the-art of 43% which uses more capable frontier models. Moreover, using current API pricing, amplifying the cheaper DeepSeek model with five samples is more cost-effective and solves more issues than paying a premium for one sample from GPT-4o or Claude 3.5 Sonnet. Interestingly, the relationship between coverage and the number of samples is often log-linear and can be modelled with an exponentiated power law, suggesting the existence of inference-time scaling laws. Finally, we find that identifying correct samples out of many generations remains an important direction for future research in domains without automatic verifiers. When solving math word problems from GSM8K and MATH, coverage with Llama-3 models grows to over 95% with 10,000 samples. However, common methods to pick correct solutions from a sample collection, such as majority voting or reward models, plateau beyond several hundred samples and fail to fully scale with the sample budget.

Sampling is a basic operation in many inference-time algorithms of large language models (LLMs). To scale up inference efficiently with a limited compute, it is crucial to find an optimal allocation for sample compute budgets: Which sampling configurations (model, temperature, language, etc.) do we use? How many samples do we generate in each configuration? We formulate these choices as a learning problem and propose OSCA, an algorithm that Optimizes Sample Compute Allocation by finding an optimal mix of different inference configurations. Our experiments show that with our learned mixed allocation, we can achieve accuracy better than the best single configuration with 128x less compute on code generation and 25x less compute on 4 reasoning tasks. OSCA is also shown to be effective in agentic workflows beyond single-turn tasks, achieving a better accuracy on SWE-Bench with 3x less compute than the default configuration. Our code and generations are released at https://github.com/LeiLiLab/OSCA.

Large language models (LLMs) have achieved significant performance gains via scaling up model sizes and/or data. However, recent evidence suggests diminishing returns from such approaches, motivating scaling the computation spent at inference time. Existing inference-time scaling methods, usually with reward models, cast the task as a search problem, which tends to be vulnerable to reward hacking as a consequence of approximation errors in reward models. In this paper, we instead cast inference-time scaling as a probabilistic inference task and leverage sampling-based techniques to explore the typical set of the state distribution of a state-space model with an approximate likelihood, rather than optimize for its mode directly. We propose a novel inference-time scaling approach by adapting particle-based Monte Carlo methods to this task. Our empirical evaluation demonstrates that our methods have a 4-16x better scaling rate over our deterministic search counterparts on various challenging mathematical reasoning tasks. Using our approach, we show that Qwen2.5-Math-1.5B-Instruct can surpass GPT-4o accuracy in only 4 rollouts, while Qwen2.5-Math-7B-Instruct scales to o1 level accuracy in only 32 rollouts. Our work not only presents an effective method to inference-time scaling, but also connects the rich literature in probabilistic inference with inference-time scaling of LLMs to develop more robust algorithms in future work. Code and further information is available at https://probabilistic-inference-scaling.github.io.

Recent innovations in generative large language models (LLMs) have made their applications and use-cases ubiquitous. This has led to large-scale deployments of these models, using complex, expensive, and power-hungry AI accelerators, most commonly GPUs. These developments make LLM inference efficiency an important challenge. Based on our extensive characterization, we find that there are two main phases during an LLM inference request: a compute-intensive prompt computation, and a memory-intensive token generation, each with distinct latency, throughput, memory, and power characteristics. Despite state-of-the-art batching and scheduling, the token generation phase underutilizes compute resources. Specifically, unlike compute-intensive prompt computation phases, token generation phases do not require the compute capability of the latest GPUs, and can be run with lower power and cost. With Splitwise, we propose splitting the two phases of a LLM inference request on to separate machines. This allows us to use hardware that is well-suited for each phase, and provision resources independently per phase. However, splitting an inference request across machines requires state transfer from the machine running prompt computation over to the machine generating tokens. We implement and optimize this state transfer using the fast back-plane interconnects available in today's GPU clusters. We use the Splitwise technique to design LLM inference clusters using the same or different types of machines for the prompt computation and token generation phases. Our clusters are optimized for three key objectives: throughput, cost, and power. In particular, we show that we can achieve 1.4x higher throughput at 20% lower cost than current designs. Alternatively, we can achieve 2.35x more throughput with the same cost and power budgets.

As large language models (LLMs) continue to grow in size, distributed inference has become increasingly important. Model-parallel strategies must now efficiently scale not only across multiple GPUs but also across multiple nodes. In this work, we present a detailed performance study of multi-node distributed inference using LLMs on GPU-based supercomputers. We conduct experiments with several state-of-the-art inference engines alongside YALIS, a research-oriented prototype engine designed for controlled experimentation. We analyze the strong-scaling behavior of different model-parallel schemes and identify key bottlenecks. Since all-reduce operations are a common performance bottleneck, we develop NVRAR, a hierarchical all-reduce algorithm based on recursive doubling with NVSHMEM. NVRAR achieves up to 1.9x-3.6x lower latency than NCCL for message sizes between 128 KB and 2 MB on HPE Slingshot and InfiniBand interconnects. Integrated into YALIS, NVRAR achieves up to a 1.72x reduction in end-to-end batch latency for the Llama 3.1 405B model in multi-node decode-heavy workloads using tensor parallelism.

Inference accounts for the majority of latency and energy consumption in large language model (LLM) deployments, often exceeding 90% of total cost. While training-time efficiency has seen extensive progress, runtime optimization remains a key bottleneck, particularly under autoregressive decoding. Existing approaches -- such as pruning, quantization, early exits, and speculative decoding -- often require retraining, architectural changes, or disrupt decoding compatibility. We introduce QuickSilver, a modular, token-level framework that enables semantic adaptivity at inference time without altering model weights or structure. QuickSilver integrates four synergistic mechanisms: (i) Dynamic Token Halting, which halts computation for tokens with converged representations; (ii) KV Cache Skipping, which selectively suppresses memory writes to reduce attention overhead; and (iii) Contextual Token Fusion, which collapses redundant tokens into shared paths to shrink sequence length. Unlike speculative decoding or MoE routing, QuickSilver operates entirely on frozen, dense models and requires no auxiliary networks. Applied to GPT-2 and Llama-2 across WikiText-103 and C4, QuickSilver achieves up to 39.6% FLOP reduction with negligible perplexity degradation (<=0.2).

Large language models (LLMs) are useful in many NLP tasks and become more capable with size, with the best open-source models having over 50 billion parameters. However, using these 50B+ models requires high-end hardware, making them inaccessible to most researchers. In this work, we investigate methods for cost-efficient inference and fine-tuning of LLMs, comparing local and distributed strategies. We observe that a large enough model (50B+) can run efficiently even on geodistributed devices in a consumer-grade network. This could allow running LLM efficiently by pooling together idle compute resources of multiple research groups and volunteers. We address two open problems: (1) how to perform inference and fine-tuning reliably if any device can disconnect abruptly and (2) how to partition LLMs between devices with uneven hardware, joining and leaving at will. In order to do that, we develop special fault-tolerant inference algorithms and load-balancing protocols that automatically assign devices to maximize the total system throughput. We showcase these algorithms in Petals - a decentralized system that runs Llama 2 (70B) and BLOOM (176B) over the Internet up to 10x faster than offloading for interactive generation. We evaluate the performance of our system in simulated conditions and a real-world setup spanning two continents.

While 4-bit quantization has emerged as a memory-optimal choice for non-reasoning models and zero-shot tasks across scales, we show that this universal prescription fails for reasoning models, where the KV cache rather than model size can dominate memory. Through systematic experiments across 1,700 inference scenarios on AIME25 and GPQA-Diamond, we find a scale-dependent trade-off: models with an effective size below 8-bit 4B parameters achieve better accuracy by allocating memory to more weights rather than longer generation, while larger models achieve better accuracy by allocating memory to longer generations. This scale threshold also determines when parallel scaling becomes memory-efficient and whether KV cache eviction outperforms KV quantization. Our findings show that memory optimization for LLMs cannot be scale-agnostic, while providing principled guidelines: for small reasoning models, prioritize model capacity over test-time compute, while for larger ones, maximize test-time compute. Our results suggest that optimizing reasoning models for deployment requires fundamentally different strategies from those established for non-reasoning models.

With the blossom of large language models (LLMs), inference efficiency becomes increasingly important. Various approximation methods are proposed to reduce the cost at inference time. Contextual Sparsity (CS) is appealing for its training-free nature and its ability to reach a higher compression ratio seemingly without quality degradation. However, after a comprehensive evaluation of contextual sparsity methods on various complex generation tasks, we find that although CS succeeds in prompt-understanding tasks, CS significantly degrades the model performance for reasoning, deduction, and knowledge-based tasks. Despite the gap in end-to-end accuracy, we observed that sparse models often share general problem-solving logic and require only a few token corrections to recover the original model performance. This paper introduces Sirius, an efficient correction mechanism, which significantly recovers CS models quality on reasoning tasks while maintaining its efficiency gain. Sirius is evaluated on 6 models with 8 difficult generation tasks in reasoning, math, and coding and shows consistent effectiveness and efficiency. Also, we carefully develop a system implementation for Sirius and show that Sirius achieves roughly 20% reduction in latency for 8B model on-chip and 35% reduction for 70B model offloading. We open-source our implementation of Sirius at https://github.com/Infini-AI-Lab/Sirius.git.

Accelerating the inference of large language models (LLMs) is an important challenge in artificial intelligence. This paper introduces distributed speculative inference (DSI), a novel distributed inference algorithm that is provably faster than speculative inference (SI) [leviathan2023fast, chen2023accelerating, miao2023specinfer] and traditional autoregressive inference (non-SI). Like other SI algorithms, DSI works on frozen LLMs, requiring no training or architectural modifications, and it preserves the target distribution. Prior studies on SI have demonstrated empirical speedups (compared to non-SI) but require a fast and accurate drafter LLM. In practice, off-the-shelf LLMs often do not have matching drafters that are sufficiently fast and accurate. We show a gap: SI gets slower than non-SI when using slower or less accurate drafters. We close this gap by proving that DSI is faster than both SI and non-SI given any drafters. By orchestrating multiple instances of the target and drafters, DSI is not only faster than SI but also supports LLMs that cannot be accelerated with SI. Our simulations show speedups of off-the-shelf LLMs in realistic settings: DSI is 1.29-1.92x faster than SI.

The large number of parameters in Pretrained Language Models enhance their performance, but also make them resource-intensive, making it challenging to deploy them on commodity hardware like a single GPU. Due to the memory and power limitations of these devices, model compression techniques are often used to decrease both the model's size and its inference latency. This usually results in a trade-off between model accuracy and efficiency. Therefore, optimizing this balance is essential for effectively deploying LLMs on commodity hardware. A significant portion of the efficiency challenge is the Feed-forward network (FFN) component, which accounts for roughly 2{3} total parameters and inference latency. In this paper, we first observe that only a few neurons of FFN module have large output norm for any input tokens, a.k.a. heavy hitters, while the others are sparsely triggered by different tokens. Based on this observation, we explicitly split the FFN into two parts according to the heavy hitters. We improve the efficiency-accuracy trade-off of existing compression methods by allocating more resource to FFN parts with heavy hitters. In practice, our method can reduce model size by 43.1\% and bring 1.25sim1.56times wall clock time speedup on different hardware with negligible accuracy drop.

Large language models (LLMs) excel in most NLP tasks but also require expensive cloud servers for deployment due to their size, while smaller models that can be deployed on lower cost (e.g., edge) devices, tend to lag behind in terms of response quality. Therefore in this work we propose a hybrid inference approach which combines their respective strengths to save cost and maintain quality. Our approach uses a router that assigns queries to the small or large model based on the predicted query difficulty and the desired quality level. The desired quality level can be tuned dynamically at test time to seamlessly trade quality for cost as per the scenario requirements. In experiments our approach allows us to make up to 40% fewer calls to the large model, with no drop in response quality.

Large Language Models (LLMs) have revolutionized natural language processing by achieving state-of-the-art results across a variety of tasks. However, the computational demands of LLM inference, including high memory consumption and slow processing speeds, pose significant challenges for real-world applications, particularly on resource-constrained devices. Efficient inference is crucial for scaling the deployment of LLMs to a broader range of platforms, including mobile and edge devices. This survey explores contemporary techniques in model compression that address these challenges by reducing the size and computational requirements of LLMs while maintaining their performance. We focus on model-level compression methods, including quantization, knowledge distillation, and pruning, as well as system-level optimizations like KV cache efficient design. Each of these methodologies offers a unique approach to optimizing LLMs, from reducing numerical precision to transferring knowledge between models and structurally simplifying neural networks. Additionally, we discuss emerging trends in system-level design that further enhance the efficiency of LLM inference. This survey aims to provide a comprehensive overview of current advancements in model compression and their potential to make LLMs more accessible and practical for diverse applications.

Large Reasoning Models (LRMs) demonstrate exceptional capability in tackling complex mathematical, logical, and coding tasks by leveraging extended Chain-of-Thought (CoT) reasoning. Test-time scaling methods, such as prolonging CoT with explicit token-level exploration, can push LRMs' accuracy boundaries, but they incur significant decoding overhead. A key inefficiency source is LRMs often generate redundant thinking CoTs, which demonstrate clear structured overthinking and underthinking patterns. Inspired by human cognitive reasoning processes and numerical optimization theories, we propose TrimR, a verifier-based, training-free, efficient framework for dynamic CoT compression to trim reasoning and enhance test-time scaling, explicitly tailored for production-level deployment. Our method employs a lightweight, pretrained, instruction-tuned verifier to detect and truncate redundant intermediate thoughts of LRMs without any LRM or verifier fine-tuning. We present both the core algorithm and asynchronous online system engineered for high-throughput industrial applications. Empirical evaluations on Ascend NPUs and vLLM show that our framework delivers substantial gains in inference efficiency under large-batch workloads. In particular, on the four MATH500, AIME24, AIME25, and GPQA benchmarks, the reasoning runtime of Pangu Pro MoE, Pangu-R-38B, QwQ-32B, and DeepSeek-R1-Distill-Qwen-32B is improved by up to 70% with negligible impact on accuracy.

The scaling of inference computation has unlocked the potential of long-context large language models (LLMs) across diverse settings. For knowledge-intensive tasks, the increased compute is often allocated to incorporate more external knowledge. However, without effectively utilizing such knowledge, solely expanding context does not always enhance performance. In this work, we investigate inference scaling for retrieval augmented generation (RAG), exploring strategies beyond simply increasing the quantity of knowledge. We focus on two inference scaling strategies: in-context learning and iterative prompting. These strategies provide additional flexibility to scale test-time computation (e.g., by increasing retrieved documents or generation steps), thereby enhancing LLMs' ability to effectively acquire and utilize contextual information. We address two key questions: (1) How does RAG performance benefit from the scaling of inference computation when optimally configured? (2) Can we predict the optimal test-time compute allocation for a given budget by modeling the relationship between RAG performance and inference parameters? Our observations reveal that increasing inference computation leads to nearly linear gains in RAG performance when optimally allocated, a relationship we describe as the inference scaling laws for RAG. Building on this, we further develop the computation allocation model to estimate RAG performance across different inference configurations. The model predicts optimal inference parameters under various computation constraints, which align closely with the experimental results. By applying these optimal configurations, we demonstrate that scaling inference compute on long-context LLMs achieves up to 58.9% gains on benchmark datasets compared to standard RAG.

Recent breakthroughs in Large Reasoning Models (LRMs) have demonstrated that extensive Chain-of-Thought (CoT) generation is critical for enabling intricate cognitive behaviors, such as self-verification and backtracking, to solve complex tasks. However, this capability often leads to ``overthinking'', where models generate redundant reasoning paths that inflate computational costs without improving accuracy. While Supervised Fine-Tuning (SFT) on reasoning traces is a standard paradigm for the 'cold start' phase, applying existing compression techniques to these traces often compromises logical coherence or incurs prohibitive sampling costs. In this paper, we introduce ConMax (Confidence-Maximizing Compression), a novel reinforcement learning framework designed to automatically compress reasoning traces while preserving essential reasoning patterns. ConMax formulates compression as a reward-driven optimization problem, training a policy to prune redundancy by maximizing a weighted combination of answer confidence for predictive fidelity and thinking confidence for reasoning validity through a frozen auxiliary LRM. Extensive experiments across five reasoning datasets demonstrate that ConMax achieves a superior efficiency-performance trade-off. Specifically, it reduces inference length by 43% over strong baselines at the cost of a mere 0.7% dip in accuracy, proving its effectiveness in generating high-quality, efficient training data for LRMs.

Reasoning models have demonstrated impressive performance in self-reflection and chain-of-thought reasoning. However, they often produce excessively long outputs, leading to prohibitively large key-value (KV) caches during inference. While chain-of-thought inference significantly improves performance on complex reasoning tasks, it can also lead to reasoning failures when deployed with existing KV cache compression approaches. To address this, we propose Redundancy-aware KV Cache Compression for Reasoning models (R-KV), a novel method specifically targeting redundant tokens in reasoning models. Our method preserves nearly 100% of the full KV cache performance using only 10% of the KV cache, substantially outperforming existing KV cache baselines, which reach only 60% of the performance. Remarkably, R-KV even achieves 105% of full KV cache performance with 16% of the KV cache. This KV-cache reduction also leads to a 90% memory saving and a 6.6X throughput over standard chain-of-thought reasoning inference. Experimental results show that R-KV consistently outperforms existing KV cache compression baselines across two mathematical reasoning datasets.

Large language model (LLM) queries are predominantly processed by frontier models in centralized cloud infrastructure. Rapidly growing demand strains this paradigm, and cloud providers struggle to scale infrastructure at pace. Two advances enable us to rethink this paradigm: small LMs (<=20B active parameters) now achieve competitive performance to frontier models on many tasks, and local accelerators (e.g., Apple M4 Max) run these models at interactive latencies. This raises the question: can local inference viably redistribute demand from centralized infrastructure? Answering this requires measuring whether local LMs can accurately answer real-world queries and whether they can do so efficiently enough to be practical on power-constrained devices (i.e., laptops). We propose intelligence per watt (IPW), task accuracy divided by unit of power, as a metric for assessing capability and efficiency of local inference across model-accelerator pairs. We conduct a large-scale empirical study across 20+ state-of-the-art local LMs, 8 accelerators, and a representative subset of LLM traffic: 1M real-world single-turn chat and reasoning queries. For each query, we measure accuracy, energy, latency, and power. Our analysis reveals 3 findings. First, local LMs can accurately answer 88.7% of single-turn chat and reasoning queries with accuracy varying by domain. Second, from 2023-2025, IPW improved 5.3x and local query coverage rose from 23.2% to 71.3%. Third, local accelerators achieve at least 1.4x lower IPW than cloud accelerators running identical models, revealing significant headroom for optimization. These findings demonstrate that local inference can meaningfully redistribute demand from centralized infrastructure, with IPW serving as the critical metric for tracking this transition. We release our IPW profiling harness for systematic intelligence-per-watt benchmarking.

Model pruning seeks to induce sparsity in a deep neural network's various connection matrices, thereby reducing the number of nonzero-valued parameters in the model. Recent reports (Han et al., 2015; Narang et al., 2017) prune deep networks at the cost of only a marginal loss in accuracy and achieve a sizable reduction in model size. This hints at the possibility that the baseline models in these experiments are perhaps severely over-parameterized at the outset and a viable alternative for model compression might be to simply reduce the number of hidden units while maintaining the model's dense connection structure, exposing a similar trade-off in model size and accuracy. We investigate these two distinct paths for model compression within the context of energy-efficient inference in resource-constrained environments and propose a new gradual pruning technique that is simple and straightforward to apply across a variety of models/datasets with minimal tuning and can be seamlessly incorporated within the training process. We compare the accuracy of large, but pruned models (large-sparse) and their smaller, but dense (small-dense) counterparts with identical memory footprint. Across a broad range of neural network architectures (deep CNNs, stacked LSTM, and seq2seq LSTM models), we find large-sparse models to consistently outperform small-dense models and achieve up to 10x reduction in number of non-zero parameters with minimal loss in accuracy.

It is commonly believed that scaling language models should commit a significant space or time cost, by increasing the parameters (parameter scaling) or output tokens (inference-time scaling). We introduce the third and more inference-efficient scaling paradigm: increasing the model's parallel computation during both training and inference time. We apply P diverse and learnable transformations to the input, execute forward passes of the model in parallel, and dynamically aggregate the P outputs. This method, namely parallel scaling (ParScale), scales parallel computation by reusing existing parameters and can be applied to any model structure, optimization procedure, data, or task. We theoretically propose a new scaling law and validate it through large-scale pre-training, which shows that a model with P parallel streams is similar to scaling the parameters by O(log P) while showing superior inference efficiency. For example, ParScale can use up to 22times less memory increase and 6times less latency increase compared to parameter scaling that achieves the same performance improvement. It can also recycle an off-the-shelf pre-trained model into a parallelly scaled one by post-training on a small amount of tokens, further reducing the training budget. The new scaling law we discovered potentially facilitates the deployment of more powerful models in low-resource scenarios, and provides an alternative perspective for the role of computation in machine learning.

We conduct experiments on the impact of increasing inference-time compute in reasoning models (specifically OpenAI o1-preview and o1-mini) on their robustness to adversarial attacks. We find that across a variety of attacks, increased inference-time compute leads to improved robustness. In many cases (with important exceptions), the fraction of model samples where the attack succeeds tends to zero as the amount of test-time compute grows. We perform no adversarial training for the tasks we study, and we increase inference-time compute by simply allowing the models to spend more compute on reasoning, independently of the form of attack. Our results suggest that inference-time compute has the potential to improve adversarial robustness for Large Language Models. We also explore new attacks directed at reasoning models, as well as settings where inference-time compute does not improve reliability, and speculate on the reasons for these as well as ways to address them.

Inference-time computation is a powerful paradigm to enhance the performance of large language models (LLMs), with Best-of-N sampling being a widely used technique. However, this method is computationally expensive, requiring both (1) an external reward model and (2) the generation of multiple samples. In this work, we introduce a new generative self-evaluation scheme designed to adaptively reduce the number of generated samples while maintaining or even improving performance. We use a generative reward model formulation, allowing the LLM to predict mid-generation the probability that restarting the generation will yield a better response. These predictions are obtained without an external reward model and can be used to decide whether or not to generate more samples, prune unpromising samples early on, or to pick the best sample. This capability is very inexpensive as it involves generating a single predefined token. Trained using a dataset constructed with real unfiltered LMSYS user prompts, Llama 3.1 8B's win rate against GPT-4 on AlpacaEval increases from 21% to 34% with 16 samples and math performance on GSM8K improves from 84% to 91%. By sampling only when the LLM determines that it is beneficial to do so and adaptively adjusting temperature annealing, we demonstrate that 74% of the improvement from using 16 samples can be achieved with only 1.2 samples on average. We further demonstrate that 50-75% of samples can be pruned early in generation with minimal degradation in performance. Overall, our methods enable more efficient and scalable compute utilization during inference for LLMs.

Inference of Large Language Models (LLMs) across computer clusters has become a focal point of research in recent times, with many acceleration techniques taking inspiration from CPU speculative execution. These techniques reduce bottlenecks associated with memory bandwidth, but also increase end-to-end latency per inference run, requiring high speculation acceptance rates to improve performance. Combined with a variable rate of acceptance across tasks, speculative inference techniques can result in reduced performance. Additionally, pipeline-parallel designs require many user requests to maintain maximum utilization. As a remedy, we propose PipeInfer, a pipelined speculative acceleration technique to reduce inter-token latency and improve system utilization for single-request scenarios while also improving tolerance to low speculation acceptance rates and low-bandwidth interconnects. PipeInfer exhibits up to a 2.15times improvement in generation speed over standard speculative inference. PipeInfer achieves its improvement through Continuous Asynchronous Speculation and Early Inference Cancellation, the former improving latency and generation speed by running single-token inference simultaneously with several speculative runs, while the latter improves speed and latency by skipping the computation of invalidated runs, even in the middle of inference.

Large reasoning models (LRMs) have exhibited the capacity of enhancing reasoning performance via internal test-time scaling. Building upon this, a promising direction is to further scale test-time compute to unlock even greater reasoning capabilities. However, as we push these scaling boundaries, systematically understanding the practical limits and achieving optimal resource allocation becomes a critical challenge. In this paper, we investigate the scaling Pareto of test-time scaling and introduce the Test-Time Scaling Performance Model (TTSPM). We theoretically analyze two fundamental paradigms for such extended scaling, parallel scaling and sequential scaling, from a probabilistic modeling perspective. Our primary contribution is the derivation of the saturation point on the scaling budget for both strategies, identifying thresholds beyond which additional computation yields diminishing returns. Remarkably, despite their distinct mechanisms, both paradigms converge to a unified mathematical structure in their upper bounds. We empirically validate our theoretical findings on challenging reasoning benchmarks, including AIME, MATH-500, and GPQA, demonstrating the practical utility of these bounds for test-time resource allocation. We hope that this work provides insights into the cost-benefit trade-offs of test-time scaling, guiding the development of more resource-efficient inference strategies for large reasoning models.

Recent research has generated hope that inference scaling could allow weaker language models to match or exceed the accuracy of stronger models, such as by repeatedly sampling solutions to a coding problem until it passes unit tests. The central thesis of this paper is that there is no free lunch for inference scaling: indefinite accuracy improvement through resampling can only be realized if the "verifier" (in this case, a set of unit tests) is perfect. When the verifier is imperfect, as it almost always is in domains such as reasoning or coding (for example, unit tests have imperfect coverage), there is a nonzero probability of false positives: incorrect solutions that pass the verifier. Resampling cannot decrease this probability, so it imposes an upper bound to the accuracy of resampling-based inference scaling even with an infinite compute budget. We find that there is a very strong correlation between the model's single-sample accuracy (i.e. accuracy without unit tests) and its false positive rate on coding benchmarks HumanEval and MBPP, whose unit tests have limited coverage. Therefore, no amount of inference scaling of weaker models can enable them to match the single-sample accuracy of a sufficiently strong model (Fig. 1a). When we consider that false positives have a negative utility compared to abstaining from producing a solution, it bends the inference scaling curve further downward. Empirically, we find that the optimal number of samples can be less than 10 under realistic assumptions (Fig. 1b). Finally, we show that beyond accuracy, false positives may have other undesirable qualities, such as poor adherence to coding style conventions.

In recent times, the emergence of Large Language Models (LLMs) has resulted in increasingly larger model size, posing challenges for inference on low-resource devices. Prior approaches have explored offloading to facilitate low-memory inference but often suffer from efficiency due to I/O bottlenecks. To achieve low-latency LLMs inference on resource-constrained devices, we introduce HeteGen, a novel approach that presents a principled framework for heterogeneous parallel computing using CPUs and GPUs. Based on this framework, HeteGen further employs heterogeneous parallel computing and asynchronous overlap for LLMs to mitigate I/O bottlenecks. Our experiments demonstrate a substantial improvement in inference speed, surpassing state-of-the-art methods by over 317% at most.

Inference-time scaling can enhance the reasoning capabilities of large language models (LLMs) on complex problems that benefit from step-by-step problem solving. Although lengthening generated scratchpads has proven effective for mathematical tasks, the broader impact of this approach on other tasks remains less clear. In this work, we investigate the benefits and limitations of scaling methods across nine state-of-the-art models and eight challenging tasks, including math and STEM reasoning, calendar planning, NP-hard problems, navigation, and spatial reasoning. We compare conventional models (e.g., GPT-4o) with models fine-tuned for inference-time scaling (e.g., o1) through evaluation protocols that involve repeated model calls, either independently or sequentially with feedback. These evaluations approximate lower and upper performance bounds and potential for future performance improvements for each model, whether through enhanced training or multi-model inference systems. Our extensive empirical analysis reveals that the advantages of inference-time scaling vary across tasks and diminish as problem complexity increases. In addition, simply using more tokens does not necessarily translate to higher accuracy in these challenging regimes. Results from multiple independent runs with conventional models using perfect verifiers show that, for some tasks, these models can achieve performance close to the average performance of today's most advanced reasoning models. However, for other tasks, a significant performance gap remains, even in very high scaling regimes. Encouragingly, all models demonstrate significant gains when inference is further scaled with perfect verifiers or strong feedback, suggesting ample potential for future improvements.

In real world, large language models (LLMs) can serve as the assistant to help users accomplish their jobs, and also support the development of advanced applications. For the wide application of LLMs, the inference efficiency is an essential concern, which has been widely studied in existing work, and numerous optimization algorithms and code libraries have been proposed to improve it. Nonetheless, users still find it challenging to compare the effectiveness of all the above methods and understand the underlying mechanisms. In this work, we perform a detailed coarse-to-fine analysis of the inference performance of various code libraries. To evaluate the overall effectiveness, we examine four usage scenarios within two practical applications. We further provide both theoretical and empirical fine-grained analyses of each module in the Transformer architecture. Our experiments yield comprehensive results that are invaluable for researchers to evaluate code libraries and improve inference strategies.

The Mixture of Experts (MoE) is an effective architecture for scaling large language models by leveraging sparse expert activation, optimizing the trade-off between performance and efficiency. However, under expert parallelism, MoE suffers from inference inefficiencies due to imbalanced token-to-expert assignment, where some experts are overloaded while others remain underutilized. This imbalance leads to poor resource utilization and increased latency, as the most burdened expert dictates the overall delay, a phenomenon we define as the \textit{Straggler Effect}. To mitigate this, we propose Capacity-Aware Inference, including two key techniques: (1) \textit{Capacity-Aware Token Drop}, which discards overloaded tokens to regulate the maximum latency of MoE, and (2) \textit{Capacity-Aware Token Reroute}, which reallocates overflowed tokens to underutilized experts, balancing the token distribution. These techniques collectively optimize both high-load and low-load expert utilization, leading to a more efficient MoE inference pipeline. Extensive experiments demonstrate the effectiveness of our methods, showing significant improvements in inference efficiency, e.g., 0.2\% average performance increase and a 1.94times inference speedup on Mixtral-8times7B-Instruct.

Recent reasoning-focused language models achieve high accuracy by generating lengthy intermediate reasoning paths before producing final answers. While this approach is effective in solving problems that require logical thinking, long reasoning paths significantly increase memory usage and throughput of token generation, limiting the practical deployment of such models. We propose Reasoning Path Compression (RPC), a training-free method that accelerates inference by leveraging the semantic sparsity of reasoning paths. RPC periodically compresses the KV cache by retaining KV cache that receive high importance score, which are computed using a selector window composed of recently generated queries. Experiments show that RPC improves generation throughput of QwQ-32B by up to 1.60times compared to the inference with full KV cache, with an accuracy drop of 1.2% on the AIME 2024 benchmark. Our findings demonstrate that semantic sparsity in reasoning traces can be effectively exploited for compression, offering a practical path toward efficient deployment of reasoning LLMs. Our code is available at https://github.com/jiwonsong-dev/ReasoningPathCompression.

Large Language Models (LLMs) have made significant strides in problem-solving by incorporating reasoning processes. However, this enhanced reasoning capability results in an increased number of output tokens during inference, leading to higher computational costs. To address this challenge, we propose TwT (Thinking without Tokens), a method that reduces inference-time costs through habitual reasoning distillation with multi-teachers' guidance, while maintaining high performance. Our approach introduces a Habitual Reasoning Distillation method, which internalizes explicit reasoning into the model's habitual behavior through a Teacher-Guided compression strategy inspired by human cognition. Additionally, we propose Dual-Criteria Rejection Sampling (DCRS), a technique that generates a high-quality and diverse distillation dataset using multiple teacher models, making our method suitable for unsupervised scenarios. Experimental results demonstrate that TwT effectively reduces inference costs while preserving superior performance, achieving up to a 13.6% improvement in accuracy with fewer output tokens compared to other distillation methods, offering a highly practical solution for efficient LLM deployment.

Large Language Models (LLMs) have demonstrated the ability to tackle increasingly complex tasks through advanced reasoning, long-form content generation, and tool use. Solving these tasks often involves long inference-time computations. In human problem solving, a common strategy to expedite work is collaboration: by dividing the problem into sub-tasks, exploring different strategies concurrently, etc. Recent research has shown that LLMs can also operate in parallel by implementing explicit cooperation frameworks, such as voting mechanisms or the explicit creation of independent sub-tasks that can be executed in parallel. However, each of these frameworks may not be suitable for all types of tasks, which can hinder their applicability. In this work, we propose a different design approach: we run LLM "workers" in parallel , allowing them to synchronize via a concurrently-updated attention cache and prompt these workers to decide how best to collaborate. Our approach allows the instances to come up with their own collaboration strategy for the problem at hand, all the while "seeing" each other's partial progress in the concurrent cache. We implement this approach via Hogwild! Inference: a parallel LLM inference engine where multiple instances of the same LLM run in parallel with the same attention cache, with "instant" access to each other's generated tokens. Hogwild! inference takes advantage of Rotary Position Embeddings (RoPE) to avoid recomputation while improving parallel hardware utilization. We find that modern reasoning-capable LLMs can perform inference with shared Key-Value cache out of the box, without additional fine-tuning.

While Large Language Models (LLMs) have achieved remarkable success in various fields, the efficiency of training and inference remains a major challenge. To address this issue, we propose SUBLLM, short for Subsampling-Upsampling-Bypass Large Language Model, an innovative architecture that extends the core decoder-only framework by incorporating subsampling, upsampling, and bypass modules. The subsampling modules are responsible for shortening the sequence, while the upsampling modules restore the sequence length, and the bypass modules enhance convergence. In comparison to LLaMA, the proposed SUBLLM exhibits significant enhancements in both training and inference speeds as well as memory usage, while maintaining competitive few-shot performance. During training, SUBLLM increases speeds by 26% and cuts memory by 10GB per GPU. In inference, it boosts speeds by up to 37% and reduces memory by 1GB per GPU. The training and inference speeds can be enhanced by 34% and 52% respectively when the context window is expanded to 8192. Our code is available at https://github.com/XiaoMi/subllm.

Large language models have become the cornerstone of natural language processing, but their use comes with substantial costs in terms of compute and memory resources. Sparsification provides a solution to alleviate these resource constraints, and recent works have shown that trained models can be sparsified post-hoc. Existing sparsification techniques face challenges as they need additional data structures and offer constrained speedup with current hardware. In this paper we present SliceGPT, a new post-training sparsification scheme which replaces each weight matrix with a smaller (dense) matrix, reducing the embedding dimension of the network. Through extensive experimentation, we show that SliceGPT can remove up to 25% of the model parameters (including embeddings) for LLAMA2-70B, OPT 66B and Phi-2 models while maintaining 99%, 99% and 90% zero-shot task performance of the dense model respectively. Our sliced models run on fewer GPUs and run faster without any additional code optimization: on 24GB consumer GPUs we reduce the total compute for inference on LLAMA2-70B to 64% of that of the dense model; on 40GB A100 GPUs we reduce it to 66%. We offer a new insight, computational invariance in transformer networks, which enables SliceGPT and we hope it will inspire and enable future avenues to reduce memory and computation demands for pre-trained models. Code is available at: https://github.com/microsoft/TransformerCompression

The application of Transformer-based large models has achieved numerous success in recent years. However, the exponential growth in the parameters of large models introduces formidable memory challenge for edge deployment. Prior works to address this challenge mainly focus on optimizing the model structure and adopting memory swapping methods. However, the former reduces the inference accuracy, and the latter raises the inference latency. This paper introduces PIPELOAD, a novel memory-efficient pipeline execution mechanism. It reduces memory usage by incorporating dynamic memory management and minimizes inference latency by employing parallel model loading. Based on PIPELOAD mechanism, we present Hermes, a framework optimized for large model inference on edge devices. We evaluate Hermes on Transformer-based models of different sizes. Our experiments illustrate that Hermes achieves up to 4.24 X increase in inference speed and 86.7% lower memory consumption than the state-of-the-art pipeline mechanism for BERT and ViT models, 2.58 X increase in inference speed and 90.3% lower memory consumption for GPT-style models.

Large Language Models (LLMs) have recently advanced the field of Automated Theorem Proving (ATP), attaining substantial performance gains through widely adopted test-time scaling strategies, notably reflective Chain-of-Thought (CoT) reasoning and increased sampling passes. However, they both introduce significant computational overhead for inference. Moreover, existing cost analyses typically regulate only the number of sampling passes, while neglecting the substantial disparities in sampling costs introduced by different scaling strategies. In this paper, we systematically compare the efficiency of different test-time scaling strategies for ATP models and demonstrate the inefficiency of the current state-of-the-art (SOTA) open-source approaches. We then investigate approaches to significantly reduce token usage and sample passes while maintaining the original performance. Specifically, we propose two complementary methods that can be integrated into a unified EconRL pipeline for amplified benefits: (1) a dynamic Chain-of-Thought (CoT) switching mechanism designed to mitigate unnecessary token consumption, and (2) Diverse parallel-scaled reinforcement learning (RL) with trainable prefixes to enhance pass rates under constrained sampling passes. Experiments on miniF2F and ProofNet demonstrate that our EconProver achieves comparable performance to baseline methods with only 12% of the computational cost. This work provides actionable insights for deploying lightweight ATP models without sacrificing performance.

We propose a new localized inference algorithm for answering marginalization queries in large graphical models with the correlation decay property. Given a query variable and a large graphical model, we define a much smaller model in a local region around the query variable in the target model so that the marginal distribution of the query variable can be accurately approximated. We introduce two approximation error bounds based on the Dobrushin's comparison theorem and apply our bounds to derive a greedy expansion algorithm that efficiently guides the selection of neighbor nodes for localized inference. We verify our theoretical bounds on various datasets and demonstrate that our localized inference algorithm can provide fast and accurate approximation for large graphical models.

Reliability on cloud providers for ASR inference to support child-centered voice-based applications is becoming challenging due to regulatory and privacy challenges. Motivated by a privacy-preserving design, this study aims to develop a lightweight & efficient Whisper ASR system capable of running on a Raspberry Pi. Upon evaluation of the MyST corpus and by examining various filtering strategies to fine-tune the `tiny.en' model, a Word Error Rate (WER) of 15.9% was achieved (11.8% filtered). A low-rank compression reduces the encoder size by 0.51M with 1.26x faster inference in GPU, with 11% relative WER increase. During inference on Pi, the compressed version required ~2 GFLOPS fewer computations. The RTF for both the models ranged between [0.23-0.41] for various input audio durations. Analyzing the RAM usage and CPU temperature showed that the PI was capable of handling both the tiny models, however it was noticed that small models initiated additional overhead/thermal throttling.

This paper presents a configurable Convolutional Neural Network Accelerator (CNNA) for a System on Chip design (SoC). The goal was to accelerate inference of different deep learning networks on an embedded SoC platform. The presented CNNA has a scalable architecture which uses High Level Synthesis (HLS) and SystemC for the hardware accelerator. It is able to accelerate any Convolutional Neural Network (CNN) exported from Python and supports a combination of convolutional, max-pooling, and fully connected layers. A training method with fixed-point quantized weights is proposed and presented in the paper. The CNNA is template-based, enabling it to scale for different targets of the Xilinx Zynq platform. This approach enables design space exploration, which makes it possible to explore several configurations of the CNNA during C- and RTL-simulation, fitting it to the desired platform and model. The CNN VGG16 was used to test the solution on a Xilinx Ultra96 board using PYNQ. The result gave a high level of accuracy in training with an auto-scaled fixed-point Q2.14 format compared to a similar floating-point model. It was able to perform inference in 2.0 seconds, while having an average power consumption of 2.63 W, which corresponds to a power efficiency of 6.0 GOPS/W.

Recent advancements in large language models (LLMs) have shifted focus toward scaling inference-time compute, improving performance without retraining the model. A common approach is to sample multiple outputs in parallel, and select one of these as the final output. However, work to date has focused on English and a handful of domains such as math and code. In contrast, we are most interested in techniques that generalize across open-ended tasks, formally verifiable tasks, and across languages. In this work, we study how to robustly scale inference-time compute for open-ended generative tasks in a multilingual, multi-task setting. Our findings show that both sampling strategy based on temperature variation and selection strategy must be adapted to account for diverse domains and varied language settings. We evaluate existing selection methods, revealing that strategies effective in English often fail to generalize across languages. We propose novel sampling and selection strategies specifically adapted for multilingual and multi-task inference scenarios, and show they yield notable gains across languages and tasks. In particular, our combined sampling and selection methods lead to an average +6.8 jump in win-rates for our 8B models on m-ArenaHard-v2.0 prompts, against proprietary models such as Gemini. At larger scale, Command-A (111B model) equipped with our methods, shows +9.0 improvement in win-rates on the same benchmark with just five samples against single-sample decoding, a substantial increase at minimal cost. Our results underscore the need for language- and task-aware approaches to inference-time compute, aiming to democratize performance improvements in underrepresented languages.

We present Klear-Reasoner, a model with long reasoning capabilities that demonstrates careful deliberation during problem solving, achieving outstanding performance across multiple benchmarks. Although there are already many excellent works related to inference models in the current community, there are still many problems with reproducing high-performance inference models due to incomplete disclosure of training details. This report provides an in-depth analysis of the reasoning model, covering the entire post-training workflow from data preparation and long Chain-of-Thought supervised fine-tuning (long CoT SFT) to reinforcement learning (RL), along with detailed ablation studies for each experimental component. For SFT data, our experiments show that a small number of high-quality data sources are more effective than a large number of diverse data sources, and that difficult samples can achieve better results without accuracy filtering. In addition, we investigate two key issues with current clipping mechanisms in RL: Clipping suppresses critical exploration signals and ignores suboptimal trajectories. To address these challenges, we propose Gradient-Preserving clipping Policy Optimization (GPPO) that gently backpropagates gradients from clipped tokens. GPPO not only enhances the model's exploration capacity but also improves its efficiency in learning from negative samples. Klear-Reasoner exhibits exceptional reasoning abilities in mathematics and programming, scoring 90.5\% on AIME 2024, 83.2\% on AIME 2025, 66.0\% on LiveCodeBench V5 and 58.1\% on LiveCodeBench V6.

Recently, large language models (LLMs) have demonstrated superior performance across various tasks by adhering to scaling laws, which significantly increase model size. However, the huge computation overhead during inference hinders the deployment in industrial applications. Many works leverage traditional compression approaches to boost model inference, but these always introduce additional training costs to restore the performance and the pruning results typically show noticeable performance drops compared to the original model when aiming for a specific level of acceleration. To address these issues, we propose a fine-grained token-wise pruning approach for the LLMs, which presents a learnable router to adaptively identify the less important tokens and skip them across model blocks to reduce computational cost during inference. To construct the router efficiently, we present a search-based sparsity scheduler for pruning sparsity allocation, a trainable router combined with our proposed four low-dimensional factors as input and three proposed losses. We conduct extensive experiments across different benchmarks on different LLMs to demonstrate the superiority of our method. Our approach achieves state-of-the-art (SOTA) pruning results, surpassing other existing pruning methods. For instance, our method outperforms BlockPruner and ShortGPT by approximately 10 points on both LLaMA2-7B and Qwen1.5-7B in accuracy retention at comparable token sparsity levels.

Large Reasoning Models (LRMs) have demonstrated remarkable performance on tasks such as mathematics and code generation. Motivated by these strengths, recent work has empirically demonstrated the effectiveness of LRMs as guard models in improving harmful query detection. However, LRMs typically generate long reasoning traces during inference, causing substantial computational overhead. In this paper, we introduce PSRT, a method that replaces the model's reasoning process with a Prefilled Safe Reasoning Trace, thereby significantly reducing the inference cost of LRMs. Concretely, PSRT prefills "safe reasoning virtual tokens" from a constructed dataset and learns over their continuous embeddings. With the aid of indicator tokens, PSRT enables harmful-query detection in a single forward pass while preserving the classification effectiveness of LRMs. We evaluate PSRT on 7 models, 13 datasets, and 8 jailbreak methods. In terms of efficiency, PSRT completely removes the overhead of generating reasoning tokens during inference. In terms of classification performance, PSRT achieves nearly identical accuracy, with only a minor average F1 drop of 0.015 across 7 models and 5 datasets.

Diffusion models have exhibited exciting capabilities in generating images and are also very promising for video creation. However, the inference speed of diffusion models is limited by the slow sampling process, restricting its use cases. The sequential denoising steps required for generating a single sample could take tens or hundreds of iterations and thus have become a significant bottleneck. This limitation is more salient for applications that are interactive in nature or require small latency. To address this challenge, we propose Partially Conditioned Patch Parallelism (PCPP) to accelerate the inference of high-resolution diffusion models. Using the fact that the difference between the images in adjacent diffusion steps is nearly zero, Patch Parallelism (PP) leverages multiple GPUs communicating asynchronously to compute patches of an image in multiple computing devices based on the entire image (all patches) in the previous diffusion step. PCPP develops PP to reduce computation in inference by conditioning only on parts of the neighboring patches in each diffusion step, which also decreases communication among computing devices. As a result, PCPP decreases the communication cost by around 70% compared to DistriFusion (the state of the art implementation of PP) and achieves 2.36sim 8.02times inference speed-up using 4sim 8 GPUs compared to 2.32sim 6.71times achieved by DistriFusion depending on the computing device configuration and resolution of generation at the cost of a possible decrease in image quality. PCPP demonstrates the potential to strike a favorable trade-off, enabling high-quality image generation with substantially reduced latency.

Recent developments in large language models have shown advantages in reallocating a notable share of computational resource from training time to inference time. However, the principles behind inference time scaling are not well understood. In this paper, we introduce an analytically tractable model of inference-time scaling: Bayesian linear regression with a reward-weighted sampler, where the reward is determined from a linear model, modeling LLM-as-a-judge scenario. We study this problem in the high-dimensional regime, where the deterministic equivalents dictate a closed-form expression for the posterior predictive mean and variance. We analyze the generalization error when training data are sampled from a teacher model. We draw k inference-time samples and select via softmax at a temperature applied to a quadratic reward. When the reward is not too different from the teacher, the generalization error decreases monotonically with increasing inference time samples k. However, the specific reward that optimizes inference-time selection generally differs from the teacher. In contrast, substantial reward misspecification induces a finite optimal k beyond which more sampling can increase the generalization error. For fixed k, there exists an optimal sampling temperature. We experimentally verify these facts in large language model inference with an additional large language model as a judge. In the "best-of-k" limit with the teacher as reward, we theoretically show that the generalization error decays as Θ(1/k^2) and determine the leading coefficient via extreme value theory. These formulas delineate domains where scaling inference-time computation is provably preferable to collecting more data. Finally, we demonstrate that when task difficulty increases, the previously mentioned advantage of inference-time compute degrades.

One of the most striking findings in modern research on large language models (LLMs) is that scaling up compute during training leads to better results. However, less attention has been given to the benefits of scaling compute during inference. This survey focuses on these inference-time approaches. We explore three areas under a unified mathematical formalism: token-level generation algorithms, meta-generation algorithms, and efficient generation. Token-level generation algorithms, often called decoding algorithms, operate by sampling a single token at a time or constructing a token-level search space and then selecting an output. These methods typically assume access to a language model's logits, next-token distributions, or probability scores. Meta-generation algorithms work on partial or full sequences, incorporating domain knowledge, enabling backtracking, and integrating external information. Efficient generation methods aim to reduce token costs and improve the speed of generation. Our survey unifies perspectives from three research communities: traditional natural language processing, modern LLMs, and machine learning systems.

Large Reasoning Models (LRMs) significantly improve the reasoning ability of Large Language Models (LLMs) by learning to reason, exhibiting promising performance in complex task-solving. However, their deliberative reasoning process leads to inefficiencies in token usage, memory consumption, and inference time. Thus, this survey provides a review of efficient inference methods designed specifically for LRMs, focusing on mitigating token inefficiency while preserving the reasoning quality. First, we introduce a taxonomy to group the recent methods into two main categories: (a) explicit compact Chain-of-Thought (CoT), which reduces tokens while keeping the explicit reasoning structure, and (b) implicit latent CoT, which encodes reasoning steps within hidden representations instead of explicit tokens. Meanwhile, we discuss their strengths and weaknesses. Then, we conduct empirical analyses on existing methods from performance and efficiency aspects. Besides, we present open challenges in this field, including human-centric controllable reasoning, trade-off between interpretability and efficiency of reasoning, ensuring safety of efficient reasoning, and broader applications of efficient reasoning. In addition, we highlight key insights for enhancing LRMs' inference efficiency via techniques such as model merging, new architectures, and agent routers. We hope this work serves as a valuable guide, helping researchers overcome challenges in this vibrant fieldhttps://github.com/yueliu1999/Awesome-Efficient-Inference-for-LRMs.

Exploiting sparsity underlying neural networks has become one of the most potential methodologies to reduce the memory footprint, I/O cost, and computation workloads during inference. And the degree of sparsity one can exploit has become higher as larger model sizes have been considered along with the trend of pre-training giant models. On the other hand, compared with quantization that has been a widely supported option, acceleration through high-degree sparsity is not supported in most computing platforms. In this work, we introduce the first commercial hardware platform supporting high-degree sparsity acceleration up to 32 times -- S4. Combined with state-of-the-art sparse pruning techniques, we demonstrate several-times practical inference speedup on S4 over mainstream inference platforms such as Nvidia T4. We also show that in practice a sparse model of larger size can achieve both higher accuracy and higher throughput on S4 than a dense model of smaller size.

Large Language Models (LLMs) with billions of parameters have drastically transformed AI applications. However, their demanding computation during inference has raised significant challenges for deployment on resource-constrained devices. Despite recent trends favoring alternative activation functions such as GELU or SiLU, known for increased computation, this study strongly advocates for reinstating ReLU activation in LLMs. We demonstrate that using the ReLU activation function has a negligible impact on convergence and performance while significantly reducing computation and weight transfer. This reduction is particularly valuable during the memory-bound inference step, where efficiency is paramount. Exploring sparsity patterns in ReLU-based LLMs, we unveil the reutilization of activated neurons for generating new tokens and leveraging these insights, we propose practical strategies to substantially reduce LLM inference computation up to three times, using ReLU activations with minimal performance trade-offs.

While neural networks have advanced the frontiers in many applications, they often come at a high computational cost. Reducing the power and latency of neural network inference is key if we want to integrate modern networks into edge devices with strict power and compute requirements. Neural network quantization is one of the most effective ways of achieving these savings but the additional noise it induces can lead to accuracy degradation. In this white paper, we introduce state-of-the-art algorithms for mitigating the impact of quantization noise on the network's performance while maintaining low-bit weights and activations. We start with a hardware motivated introduction to quantization and then consider two main classes of algorithms: Post-Training Quantization (PTQ) and Quantization-Aware-Training (QAT). PTQ requires no re-training or labelled data and is thus a lightweight push-button approach to quantization. In most cases, PTQ is sufficient for achieving 8-bit quantization with close to floating-point accuracy. QAT requires fine-tuning and access to labeled training data but enables lower bit quantization with competitive results. For both solutions, we provide tested pipelines based on existing literature and extensive experimentation that lead to state-of-the-art performance for common deep learning models and tasks.

Training-free acceleration has emerged as an advanced research area in video generation based on diffusion models. The redundancy of latents in diffusion model inference provides a natural entry point for acceleration. In this paper, we decompose the inference process into the encoding, denoising, and decoding stages, and observe that cache-based acceleration methods often lead to substantial memory surges in the latter two stages. To address this problem, we analyze the characteristics of inference across different stages and propose stage-specific strategies for reducing memory consumption: 1) Asynchronous Cache Swapping. 2) Feature chunk. 3) Slicing latents to decode. At the same time, we ensure that the time overhead introduced by these three strategies remains lower than the acceleration gains themselves. Compared with the baseline, our approach achieves faster inference speed and lower memory usage, while maintaining quality degradation within an acceptable range. The Code is available at https://github.com/NKUShaw/LightCache .

The simultaneous rise of machine learning as a service and concerns over user privacy have increasingly motivated the need for private inference (PI). While recent work demonstrates PI is possible using cryptographic primitives, the computational overheads render it impractical. The community is largely unprepared to address these overheads, as the source of slowdown in PI stems from the ReLU operator whereas optimizations for plaintext inference focus on optimizing FLOPs. In this paper we re-think the ReLU computation and propose optimizations for PI tailored to properties of neural networks. Specifically, we reformulate ReLU as an approximate sign test and introduce a novel truncation method for the sign test that significantly reduces the cost per ReLU. These optimizations result in a specific type of stochastic ReLU. The key observation is that the stochastic fault behavior is well suited for the fault-tolerant properties of neural network inference. Thus, we provide significant savings without impacting accuracy. We collectively call the optimizations Circa and demonstrate improvements of up to 4.7x storage and 3x runtime over baseline implementations; we further show that Circa can be used on top of recent PI optimizations to obtain 1.8x additional speedup.

Generative models have made significant impacts across various domains, largely due to their ability to scale during training by increasing data, computational resources, and model size, a phenomenon characterized by the scaling laws. Recent research has begun to explore inference-time scaling behavior in Large Language Models (LLMs), revealing how performance can further improve with additional computation during inference. Unlike LLMs, diffusion models inherently possess the flexibility to adjust inference-time computation via the number of denoising steps, although the performance gains typically flatten after a few dozen. In this work, we explore the inference-time scaling behavior of diffusion models beyond increasing denoising steps and investigate how the generation performance can further improve with increased computation. Specifically, we consider a search problem aimed at identifying better noises for the diffusion sampling process. We structure the design space along two axes: the verifiers used to provide feedback, and the algorithms used to find better noise candidates. Through extensive experiments on class-conditioned and text-conditioned image generation benchmarks, our findings reveal that increasing inference-time compute leads to substantial improvements in the quality of samples generated by diffusion models, and with the complicated nature of images, combinations of the components in the framework can be specifically chosen to conform with different application scenario.

The growing demand for on-device large language model (LLM) inference is driving interest in deploying lightweight, cost-effective AI solutions on edge hardware. Single-board computers (SBCs) such as the Raspberry Pi and Orange Pi offer a promising platform for localized, privacy-preserving inference-but remain underexplored in the context of LLM workloads. In this work, we benchmark the performance of 25 quantized open-source LLMs across three SBCs-Raspberry Pi 4, Raspberry Pi 5, and Orange Pi 5 Pro-using two inference runtimes: Ollama and Llamafile. We evaluate generation throughput, memory usage, and power consumption under varying CPU configurations, using multiple prompt types to simulate realistic workloads. Our results show that SBCs can reliably support models up to 1.5B parameters, with Llamafile achieving up to 4x higher throughput and 30-40% lower power usage than Ollama. We identify architecture-specific bottlenecks, highlight runtime-level trade-offs, and provide practical deployment recommendations. This study offers the first broad evaluation of LLM inference on SBCs, bridging the gap between high-performance language models and affordable edge computing.

Transformer-based language models have become the standard approach to solving natural language processing tasks. However, industry adoption usually requires the maximum throughput to comply with certain latency constraints that prevents Transformer models from being used in production. To address this gap, model compression techniques such as quantization and pruning may be used to improve inference efficiency. However, these compression techniques require specialized software to apply and deploy at scale. In this work, we propose a new pipeline for creating and running Fast Transformer models on CPUs, utilizing hardware-aware pruning, knowledge distillation, quantization, and our own Transformer inference runtime engine with optimized kernels for sparse and quantized operators. We demonstrate the efficiency of our pipeline by creating a Fast DistilBERT model showing minimal accuracy loss on the question-answering SQuADv1.1 benchmark, and throughput results under typical production constraints and environments. Our results outperform existing state-of-the-art Neural Magic's DeepSparse runtime performance by up to 50% and up to 4.1x performance speedup over ONNX Runtime. Source code is publicly available at https://github.com/intel/intel-extension-for-transformers.

Recently, large-scale transformer-based models have been proven to be effective over various tasks across many domains. Nevertheless, applying them in industrial production requires tedious and heavy works to reduce inference costs. To fill such a gap, we introduce a scalable inference solution: Easy and Efficient Transformer (EET), including a series of transformer inference optimization at the algorithm and implementation levels. First, we design highly optimized kernels for long inputs and large hidden sizes. Second, we propose a flexible CUDA memory manager to reduce the memory footprint when deploying a large model. Compared with the state-of-the-art transformer inference library (Faster Transformer v4.0), EET can achieve an average of 1.40-4.20x speedup on the transformer decoder layer with an A100 GPU

Enabling LLMs to improve their outputs by using more test-time computation is a critical step towards building generally self-improving agents that can operate on open-ended natural language. In this paper, we study the scaling of inference-time computation in LLMs, with a focus on answering the question: if an LLM is allowed to use a fixed but non-trivial amount of inference-time compute, how much can it improve its performance on a challenging prompt? Answering this question has implications not only on the achievable performance of LLMs, but also on the future of LLM pretraining and how one should tradeoff inference-time and pre-training compute. Despite its importance, little research attempted to understand the scaling behaviors of various test-time inference methods. Moreover, current work largely provides negative results for a number of these strategies. In this work, we analyze two primary mechanisms to scale test-time computation: (1) searching against dense, process-based verifier reward models; and (2) updating the model's distribution over a response adaptively, given the prompt at test time. We find that in both cases, the effectiveness of different approaches to scaling test-time compute critically varies depending on the difficulty of the prompt. This observation motivates applying a "compute-optimal" scaling strategy, which acts to most effectively allocate test-time compute adaptively per prompt. Using this compute-optimal strategy, we can improve the efficiency of test-time compute scaling by more than 4x compared to a best-of-N baseline. Additionally, in a FLOPs-matched evaluation, we find that on problems where a smaller base model attains somewhat non-trivial success rates, test-time compute can be used to outperform a 14x larger model.

Large Language Models (LLMs) have long held sway in the realms of artificial intelligence research. Numerous efficient techniques, including weight pruning, quantization, and distillation, have been embraced to compress LLMs, targeting memory reduction and inference acceleration, which underscore the redundancy in LLMs. However, most model compression techniques concentrate on weight optimization, overlooking the exploration of optimal architectures. Besides, traditional architecture search methods, limited by the elevated complexity with extensive parameters, struggle to demonstrate their effectiveness on LLMs. In this paper, we propose a training-free architecture search framework to identify optimal subnets that preserve the fundamental strengths of the original LLMs while achieving inference acceleration. Furthermore, after generating subnets that inherit specific weights from the original LLMs, we introduce a reformation algorithm that utilizes the omitted weights to rectify the inherited weights with a small amount of calibration data. Compared with SOTA training-free structured pruning works that can generate smaller networks, our method demonstrates superior performance across standard benchmarks. Furthermore, our generated subnets can directly reduce the usage of GPU memory and achieve inference acceleration.

Despite their remarkable success in complex tasks propelling widespread adoption, large language-model-based agents still face critical deployment challenges due to prohibitive latency and inference costs. While recent work has explored various methods to accelerate inference, existing approaches suffer from significant limitations: they either fail to preserve performance fidelity, require extensive offline training of router modules, or incur excessive operational costs. Moreover, they provide minimal user control over the tradeoff between acceleration and other performance metrics. To address these gaps, we introduce Dynamic Speculative Planning (DSP), an asynchronous online reinforcement learning framework that provides lossless acceleration with substantially reduced costs without requiring additional pre-deployment preparation. DSP explicitly optimizes a joint objective balancing end-to-end latency against dollar cost, allowing practitioners to adjust a single parameter that steers the system toward faster responses, cheaper operation, or any point along this continuum. Experiments on two standard agent benchmarks demonstrate that DSP achieves comparable efficiency to the fastest lossless acceleration method while reducing total cost by 30% and unnecessary cost up to 60%. Our code and data are available through https://github.com/guanyilin428/Dynamic-Speculative-Planning.

Machine learning inference occurs at a massive scale, yet its environmental impact remains poorly quantified, especially on low-resource hardware. We present ML-EcoLyzer, a cross-framework tool for measuring the carbon, energy, thermal, and water costs of inference across CPUs, consumer GPUs, and datacenter accelerators. The tool supports both classical and modern models, applying adaptive monitoring and hardware-aware evaluation. We introduce the Environmental Sustainability Score (ESS), which quantifies the number of effective parameters served per gram of CO_2 emitted. Our evaluation covers over 1,900 inference configurations, spanning diverse model architectures, task modalities (text, vision, audio, tabular), hardware types, and precision levels. These rigorous and reliable measurements demonstrate that quantization enhances ESS, huge accelerators can be inefficient for lightweight applications, and even small models may incur significant costs when implemented suboptimally. ML-EcoLyzer sets a standard for sustainability-conscious model selection and offers an extensive empirical evaluation of environmental costs during inference.

Mixture-of-Experts (MoE) models have become a key approach for scaling large language models efficiently by activating only a subset of experts during training and inference. Typically, the number of activated experts presents a trade-off: fewer experts reduce computational costs, while more experts improve performance. Recent studies reveal that not all activated experts contribute equally to model performance, with some providing minimal utility, particularly when finetuning pretrained MoE models for specialized downstream tasks. The co-existence of significant and redundant parameters in experts provides us an opportunity to reduce the number of activated experts while maintaining model performance. In this work, we propose the concept of compressed experts, lightweight modules that serve as compact representations of full experts. Our approach preserves the most important experts while replacing other auxiliary activated experts with compressed experts. The reduction of active parameters significantly lowers inference costs while achieving comparable performance. Extensive experiments on models including Phi-MoE and OLMoE demonstrate that compressed experts recover over 90% of full expert performance across various tasks while reducing more than 30% active parameters and saving 20% in inference costs. This approach enables efficient deployment of MoE models in resource-constrained settings and facilitates scaling to larger models with manageable overhead. Our code is available at https://github.com/yifei-he/Compressed-Experts.

Chain-of-Thought significantly enhances a model's reasoning capability, but it also comes with a considerable increase in inference costs due to long chains. With the observation that the reasoning path can be easily compressed under easy tasks but struggle on hard tasks, we explore the feasibility of elastically controlling the length of reasoning paths with only one model, thereby reducing the inference overhead of reasoning models dynamically based on task difficulty. We introduce a new tuning and inference strategy named CoT-Valve, designed to allow models to generate reasoning chains of varying lengths. To achieve this, we propose to identify a direction in the parameter space that, when manipulated, can effectively control the length of generated CoT. Moreover, we show that this property is valuable for compressing the reasoning chain. We construct datasets with chains from long to short for the same questions and explore two enhanced strategies for CoT-Valve: (1) a precise length-compressible CoT tuning method, and (2) a progressive chain length compression approach. Our experiments show that CoT-Valve successfully enables controllability and compressibility of the chain and shows better performance than the prompt-based control. We applied this method to QwQ-32B-Preview, reducing reasoning chains on GSM8K from 741 to 225 tokens with a minor performance drop (95.07% to 94.92%) and on AIME from 6827 to 4629 tokens, with only one additional incorrect answer.

In this paper, we present a novel method that reduces model inference latency during distributed deployment of Large Language Models (LLMs). Our contribution is an optimized inference deployment scheme that address the current limitations of state-of-the-art quantization kernels when used in conjunction with Tensor Parallel (TP). Our method preserves data locality in GPU memory access patterns and exploits a priori knowledge of TP to reduce global communication. We demonstrate an up to 1.81x speedup over existing methods for Llama-70B and up to 1.78x speedup for IBM WatsonX's Granite-20B MLP layer problem sizes on A100 and H100 NVIDIA DGX Systems for a variety of TP settings.

Large language models have been widely adopted across different tasks, but their auto-regressive generation nature often leads to inefficient resource utilization during inference. While batching is commonly used to increase throughput, performance gains plateau beyond a certain batch size, especially with smaller models, a phenomenon that existing literature typically explains as a shift to the compute-bound regime. In this paper, through an in-depth GPU-level analysis, we reveal that large-batch inference remains memory-bound, with most GPU compute capabilities underutilized due to DRAM bandwidth saturation as the primary bottleneck. To address this, we propose a Batching Configuration Advisor (BCA) that optimizes memory allocation, reducing GPU memory requirements with minimal impact on throughput. The freed memory and underutilized GPU compute capabilities can then be leveraged by concurrent workloads. Specifically, we use model replication to improve serving throughput and GPU utilization. Our findings challenge conventional assumptions about LLM inference, offering new insights and practical strategies for improving resource utilization, particularly for smaller language models.

The recent rise of privacy concerns has led researchers to devise methods for private neural inference -- where inferences are made directly on encrypted data, never seeing inputs. The primary challenge facing private inference is that computing on encrypted data levies an impractically-high latency penalty, stemming mostly from non-linear operators like ReLU. Enabling practical and private inference requires new optimization methods that minimize network ReLU counts while preserving accuracy. This paper proposes DeepReDuce: a set of optimizations for the judicious removal of ReLUs to reduce private inference latency. The key insight is that not all ReLUs contribute equally to accuracy. We leverage this insight to drop, or remove, ReLUs from classic networks to significantly reduce inference latency and maintain high accuracy. Given a target network, DeepReDuce outputs a Pareto frontier of networks that tradeoff the number of ReLUs and accuracy. Compared to the state-of-the-art for private inference DeepReDuce improves accuracy and reduces ReLU count by up to 3.5% (iso-ReLU count) and 3.5times (iso-accuracy), respectively.

The growing demand for efficient Large Language Model (LLM) inference requires a holistic optimization on algorithms, systems, and hardware. However, very few works have fundamentally changed the generation pattern: each token needs one forward pass and one KV cache. This can be sub-optimal because we found that LLMs are extremely capable of self-identifying the exact dose of information that a single KV cache can store, and many tokens can be generated confidently without global context. Based on this insight, we introduce HAMburger, a Hierarchically Auto-regressive Model that redefines resource allocation in LLMs by moving beyond uniform computation and storage per token during inference. Stacking a compositional embedder and a micro-step decoder in between a base LLM, HAMburger smashes multiple tokens into a single KV and generates several tokens per step. Additionally, HAMburger functions as a speculative decoding framework where it can blindly trust self-drafted tokens. As a result, HAMburger shifts the growth of KV cache and forward FLOPs from linear to sub-linear with respect to output length, and adjusts its inference speed based on query perplexity and output structure. Extensive evaluations show that HAMburger reduces the KV cache computation by up to 2times and achieves up to 2times TPS, while maintaining quality in both short- and long-context tasks. Our method explores an extremely challenging inference regime that requires both computation- and memory-efficiency with a hardware-agnostic design.

We study the problem of efficient generative inference for Transformer models, in one of its most challenging settings: large deep models, with tight latency targets and long sequence lengths. Better understanding of the engineering tradeoffs for inference for large Transformer-based models is important as use cases of these models are growing rapidly throughout application areas. We develop a simple analytical model for inference efficiency to select the best multi-dimensional partitioning techniques optimized for TPU v4 slices based on the application requirements. We combine these with a suite of low-level optimizations to achieve a new Pareto frontier on the latency and model FLOPS utilization (MFU) tradeoffs on 500B+ parameter models that outperforms the FasterTransformer suite of benchmarks. We further show that with appropriate partitioning, the lower memory requirements of multiquery attention (i.e. multiple query heads share single key/value head) enables scaling up to 32x larger context lengths. Finally, we achieve a low-batch-size latency of 29ms per token during generation (using int8 weight quantization) and a 76% MFU during large-batch-size processing of input tokens, while supporting a long 2048-token context length on the PaLM 540B parameter model.

Scaling test-time compute has emerged as a key strategy for enhancing the reasoning capabilities of large language models (LLMs), particularly in tasks like mathematical problem-solving. A traditional approach, Self-Consistency (SC), generates multiple solutions to a problem and selects the most common answer via majority voting. Another common method involves scoring each solution with a reward model (verifier) and choosing the best one. Recent advancements in Generative Reward Models (GenRM) reframe verification as a next-token prediction task, enabling inference-time scaling along a new axis. Specifically, GenRM generates multiple verification chains-of-thought to score each solution. Under a limited inference budget, this introduces a fundamental trade-off: should you spend the budget on scaling solutions via SC or generate fewer solutions and allocate compute to verification via GenRM? To address this, we evaluate GenRM against SC under a fixed inference budget. Interestingly, we find that SC is more compute-efficient than GenRM for most practical inference budgets across diverse models and datasets. For instance, GenRM first matches SC after consuming up to 8x the inference compute and requires significantly more compute to outperform it. Furthermore, we derive inference scaling laws for the GenRM paradigm, revealing that compute-optimal inference favors scaling solution generation more aggressively than scaling the number of verifications. Our work provides practical guidance on optimizing test-time scaling by balancing solution generation and verification. The code is available at https://github.com/nishadsinghi/sc-genrm-scaling.

Large Language Models (LLMs) are becoming integral to daily life, showcasing their vast potential across various Natural Language Processing (NLP) tasks. Beyond NLP, LLMs are increasingly used in software development tasks, such as code completion, modification, bug fixing, and code translation. Software engineers widely use tools like GitHub Copilot and Amazon Q, streamlining workflows and automating tasks with high accuracy. While the resource and energy intensity of LLM training is often highlighted, inference can be even more resource-intensive over time, as it's a continuous process with a high number of invocations. Therefore, developing resource-efficient alternatives for LLM inference is crucial for sustainability. This work proposes GREEN-CODE, a framework for energy-aware code generation in LLMs. GREEN-CODE performs dynamic early exit during LLM inference. We train a Reinforcement Learning (RL) agent that learns to balance the trade-offs between accuracy, latency, and energy consumption. Our approach is evaluated on two open-source LLMs, Llama 3.2 3B and OPT 2.7B, using the JavaCorpus and PY150 datasets. Results show that our method reduces the energy consumption between 23-50 % on average for code generation tasks without significantly affecting accuracy.

The introduction of ChatGPT has led to a significant increase in the utilization of Large Language Models (LLMs) for addressing downstream tasks. There's an increasing focus on cost-efficient training and deployment within this context. Low-cost training and deployment of LLMs represent the future development trend. This paper reviews the evolution of large language model training techniques and inference deployment technologies aligned with this emerging trend. The discussion on training includes various aspects, including data preprocessing, training architecture, pre-training tasks, parallel training, and relevant content related to model fine-tuning. On the inference side, the paper covers topics such as model compression, parallel computation, memory scheduling, and structural optimization. It also explores LLMs' utilization and provides insights into their future development.

Large Language Models (LLMs) are increasingly relied upon for solving complex reasoning tasks in domains such as mathematics, logic, and multi-step question answering. A growing line of work seeks to improve reasoning quality by scaling inference time compute particularly through Process Reward Models (PRMs), used to reward the reasoning at intermediate steps. While effective, these methods introduce substantial computational overhead, especially when generating large numbers of solutions in parallel. In this paper, we investigate whether PRMs can be used mid-generation to provide early signals that enable the rejection of suboptimal candidates before full generation of step is complete. We introduce the hypothesis that PRMs are also Partial Reward Models, meaning that the scores they assign to partially completed reasoning step are predictive of final output quality. This allows for principled early rejection based on intermediate token-level signals. We support this hypothesis both theoretically, by proving that the risk of discarding optimal beams decreases exponentially with generation length and empirically, by demonstrating a strong correlation between partial and final rewards across multiple reward models. On math reasoning benchmarks, our method achieves up to 1.4times-9times reduction in inference FLOPs without degrading final performance. These results suggest that early rejection is a powerful mechanism for improving the compute-efficiency of reasoning in LLMs.

Large Reasoning Models (LRMs) perform strongly in complex reasoning tasks via Chain-of-Thought (CoT) prompting, but often suffer from verbose outputs caused by redundant content, increasing computational overhead, and degrading user experience. Existing compression methods either operate post-hoc pruning, risking disruption to reasoning coherence, or rely on sampling-based selection, which fails to intervene effectively during generation. In this work, we introduce a confidence-guided perspective to explain the emergence of redundant reflection in LRMs, identifying two key patterns: Confidence Deficit, where the model reconsiders correct steps due to low internal confidence, and Termination Delay, where reasoning continues even after reaching a confident answer. Based on this analysis, we propose ConCISE (Confidence-guided Compression In Step-by-step Efficient Reasoning), a framework that simplifies reasoning chains by reinforcing the model's confidence during inference, thus preventing the generation of redundant reflection steps. It integrates Confidence Injection to stabilize intermediate steps and Early Stopping to terminate reasoning when confidence is sufficient. Extensive experiments demonstrate that fine-tuning LRMs on ConCISE-generated data yields significantly shorter outputs, reducing length by up to approximately 50% under SimPO, while maintaining high task accuracy. ConCISE consistently outperforms existing baselines across multiple reasoning benchmarks.

Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing tasks. Exploiting the heterogeneous capabilities of edge LLMs is crucial for diverse emerging applications, as it enables greater cost-effectiveness and reduced latency. In this work, we introduce Mixture-of-Edge-Experts (MoE^2), a novel collaborative inference framework for edge LLMs. We formulate the joint gating and expert selection problem to optimize inference performance under energy and latency constraints. Unlike conventional MoE problems, LLM expert selection is significantly more challenging due to the combinatorial nature and the heterogeneity of edge LLMs across various attributes. To this end, we propose a two-level expert selection mechanism through which we uncover an optimality-preserving property of gating parameters across expert selections. This property enables the decomposition of the training and selection processes, significantly reducing complexity. Furthermore, we leverage the objective's monotonicity and design a discrete monotonic optimization algorithm for optimal expert selection. We implement edge servers with NVIDIA Jetson AGX Orins and NVIDIA RTX 4090 GPUs, and perform extensive experiments. Our results validate that performance improvements of various LLM models and show that our MoE^2 method can achieve optimal trade-offs among different delay and energy budgets, and outperforms baselines under various system resource constraints.

The Large Language Model (LLM) is widely employed for tasks such as intelligent assistants, text summarization, translation, and multi-modality on mobile phones. However, the current methods for on-device LLM deployment maintain slow inference speed, which causes poor user experience. To facilitate high-efficiency LLM deployment on device GPUs, we propose four optimization techniques: (a) a symbolic expression-based approach to support dynamic shape model inference; (b) operator optimizations and execution priority setting to enhance inference speed and reduce phone lagging; (c) an FP4 quantization method termed M0E4 to reduce dequantization overhead; (d) a sub-tensor-based technique to eliminate the need for copying KV cache after LLM inference. Furthermore, we implement these methods in our mobile inference engine, Transformer-Lite, which is compatible with both Qualcomm and MTK processors. We evaluated Transformer-Lite's performance using LLMs with varied architectures and parameters ranging from 2B to 14B. Specifically, we achieved prefill and decoding speeds of 121 token/s and 14 token/s for ChatGLM2 6B, and 330 token/s and 30 token/s for smaller Gemma 2B, respectively. Compared with CPU-based FastLLM and GPU-based MLC-LLM, our engine attains over 10x speedup for the prefill speed and 2~3x speedup for the decoding speed.

Large language models (LLMs) power many state-of-the-art systems in natural language processing. However, these models are extremely computationally expensive, even at inference time, raising the natural question: when is the extra cost of deploying a larger model worth the anticipated boost in capabilities? Better understanding this tradeoff fundamentally could benefit from an inference efficiency metric that is both (i) easily comparable across models from different providers, and (ii) representative of the true cost of running queries in an isolated performance environment. Unfortunately, access to LLMs today is largely restricted to black-box text generation APIs and raw runtimes measured through this interface do not satisfy these desiderata: model providers can apply various software and hardware optimizations orthogonal to the model, and models served on shared infrastructure are susceptible to performance contention. To circumvent these problems, we propose a new metric for comparing inference efficiency across models. This metric puts models on equal footing as though they were served (i) on uniform hardware and software, and (ii) without performance contention. We call this metric the idealized runtime, and we propose a methodology to efficiently estimate this metric for autoregressive Transformer models. We also propose cost-aware variants that incorporate the number of accelerators needed to serve the model. Using these metrics, we compare ten state-of-the-art LLMs to provide the first analysis of inference efficiency-capability tradeoffs; we make several observations from this analysis, including the fact that the superior inference runtime performance of certain APIs is often a byproduct of optimizations within the API rather than the underlying model. Our methodology also facilitates the efficient comparison of different software and hardware stacks.

We provide theoretical and empirical evidence that using tighter evidence lower bounds (ELBOs) can be detrimental to the process of learning an inference network by reducing the signal-to-noise ratio of the gradient estimator. Our results call into question common implicit assumptions that tighter ELBOs are better variational objectives for simultaneous model learning and inference amortization schemes. Based on our insights, we introduce three new algorithms: the partially importance weighted auto-encoder (PIWAE), the multiply importance weighted auto-encoder (MIWAE), and the combination importance weighted auto-encoder (CIWAE), each of which includes the standard importance weighted auto-encoder (IWAE) as a special case. We show that each can deliver improvements over IWAE, even when performance is measured by the IWAE target itself. Furthermore, our results suggest that PIWAE may be able to deliver simultaneous improvements in the training of both the inference and generative networks.

While frontier large language models (LLMs) continue to push capability boundaries, their deployment remains confined to GPU-powered cloud infrastructure. We challenge this paradigm with SmallThinker, a family of LLMs natively designed - not adapted - for the unique constraints of local devices: weak computational power, limited memory, and slow storage. Unlike traditional approaches that mainly compress existing models built for clouds, we architect SmallThinker from the ground up to thrive within these limitations. Our innovation lies in a deployment-aware architecture that transforms constraints into design principles. First, We introduce a two-level sparse structure combining fine-grained Mixture-of-Experts (MoE) with sparse feed-forward networks, drastically reducing computational demands without sacrificing model capacity. Second, to conquer the I/O bottleneck of slow storage, we design a pre-attention router that enables our co-designed inference engine to prefetch expert parameters from storage while computing attention, effectively hiding storage latency that would otherwise cripple on-device inference. Third, for memory efficiency, we utilize NoPE-RoPE hybrid sparse attention mechanism to slash KV cache requirements. We release SmallThinker-4B-A0.6B and SmallThinker-21B-A3B, which achieve state-of-the-art performance scores and even outperform larger LLMs. Remarkably, our co-designed system mostly eliminates the need for expensive GPU hardware: with Q4_0 quantization, both models exceed 20 tokens/s on ordinary consumer CPUs, while consuming only 1GB and 8GB of memory respectively. SmallThinker is publicly available at hf.co/PowerInfer/SmallThinker-4BA0.6B-Instruct and hf.co/PowerInfer/SmallThinker-21BA3B-Instruct.

Progress on training and architecture strategies has enabled LLMs with millions of tokens in context length. However, empirical evidence suggests that such long-context LLMs can consume far more text than they can reliably use. On the other hand, it has been shown that inference-time compute can be used to scale performance of LLMs, often by generating thinking tokens, on challenging tasks involving multi-step reasoning. Through controlled experiments on sandbox long-context tasks, we find that such inference-time strategies show rapidly diminishing returns and fail at long context. We attribute these failures to score dilution, a phenomenon inherent to static self-attention. Further, we show that current inference-time strategies cannot retrieve relevant long-context signals under certain conditions. We propose a simple method that, through targeted gradient updates on the given context, provably overcomes limitations of static self-attention. We find that this shift in how inference-time compute is spent leads to consistently large performance improvements across models and long-context benchmarks. Our method leads to large 12.6 and 14.1 percentage point improvements for Qwen3-4B on average across subsets of LongBench-v2 and ZeroScrolls benchmarks. The takeaway is practical: for long context, a small amount of context-specific training is a better use of inference compute than current inference-time scaling strategies like producing more thinking tokens.

Large reasoning models achieve strong performance on complex tasks by generating extended chains of thought, but they often "overthink": continuing to reason long after they have enough information to answer correctly. This wastes inference-time compute and can hurt accuracy. Existing attempts to stop early either manipulate decoding with extra sampling and heuristics, rely on auxiliary verifier models, or operate only as post-hoc analysis pipelines without formal guarantees. We introduce LYNX, an online early-exit mechanism that turns a model's own hidden-state awareness into confidence-controlled stopping decisions. LYNX attaches exit decisions to naturally occurring reasoning cues (e.g., "hmm", "wait") during generation, trains a lightweight probe on hidden states at those cue tokens using supervision from forced exits, and wraps the resulting scores in split conformal prediction to obtain distribution-free control over premature exits. Crucially, we train and calibrate this probe once on a generic mathematical corpus and reuse it unchanged across benchmarks, decoding temperatures, and even non-mathematical tasks. Across three model families spanning 1.5B to 32B parameters, a single mathematically trained probe per base model yields strong accuracy--efficiency tradeoffs. On GSM8K, LYNX matches or improves baseline accuracy while reducing tokens by 40--65\%; on MATH-500 it improves accuracy by up to 12 points with roughly 35--60\% fewer tokens; on AIME 2024 it recovers baseline accuracy with more than 50\% token savings; and on CommonsenseQA, a non-math benchmark, it transfers zero-shot with modest accuracy gains and up to 70\% fewer tokens. Compared to state-of-the-art early-exit methods, LYNX offers competitive or superior Pareto frontiers while remaining fully online, requiring no proxy models at inference, and providing explicit, user-tunable confidence guarantees.

Matrix multiplication (MatMul) typically dominates the overall computational cost of large language models (LLMs). This cost only grows as LLMs scale to larger embedding dimensions and context lengths. In this work, we show that MatMul operations can be completely eliminated from LLMs while maintaining strong performance at billion-parameter scales. Our experiments show that our proposed MatMul-free models achieve performance on-par with state-of-the-art Transformers that require far more memory during inference at a scale up to at least 2.7B parameters. We investigate the scaling laws and find that the performance gap between our MatMul-free models and full precision Transformers narrows as the model size increases. We also provide a GPU-efficient implementation of this model which reduces memory usage by up to 61% over an unoptimized baseline during training. By utilizing an optimized kernel during inference, our model's memory consumption can be reduced by more than 10x compared to unoptimized models. To properly quantify the efficiency of our architecture, we build a custom hardware solution on an FPGA which exploits lightweight operations beyond what GPUs are capable of. We processed billion-parameter scale models at 13W beyond human readable throughput, moving LLMs closer to brain-like efficiency. This work not only shows how far LLMs can be stripped back while still performing effectively, but also points at the types of operations future accelerators should be optimized for in processing the next generation of lightweight LLMs. Our code implementation is available at https://github.com/ridgerchu/matmulfreellm.

We introduce stochastic activations. This novel strategy randomly selects between several non-linear functions in the feed-forward layer of a large language model. In particular, we choose between SILU or RELU depending on a Bernoulli draw. This strategy circumvents the optimization problem associated with RELU, namely, the constant shape for negative inputs that prevents the gradient flow. We leverage this strategy in two ways: (1) We use stochastic activations during pre-training and fine-tune the model with RELU, which is used at inference time to provide sparse latent vectors. This reduces the inference FLOPs and translates into a significant speedup in the CPU. Interestingly, this leads to much better results than training from scratch with the RELU activation function. (2) We evaluate stochastic activations for generation. This strategy performs reasonably well: it is only slightly inferior to the best deterministic non-linearity, namely SILU combined with temperature scaling. This offers an alternative to existing strategies by providing a controlled way to increase the diversity of the generated text.

Techniques that enhance inference through increased computation at test-time have recently gained attention. In this survey, we investigate the current state of LLM Inference-Time Self-Improvement from three different perspectives: Independent Self-improvement, focusing on enhancements via decoding or sampling methods; Context-Aware Self-Improvement, leveraging additional context or datastore; and Model-Aided Self-Improvement, achieving improvement through model collaboration. We provide a comprehensive review of recent relevant studies, contribute an in-depth taxonomy, and discuss challenges and limitations, offering insights for future research.

Large language models (LLMs) have revolutionized the field of AI, demonstrating unprecedented capacity across various tasks. However, the inference process for LLMs comes with significant computational costs. In this paper, we propose an efficient LLM inference pipeline that harnesses the power of LLMs. Our approach begins by tapping into the potential of LLMs to accurately perceive and predict the response length with minimal overhead. By leveraging this information, we introduce an efficient sequence scheduling technique that groups queries with similar response lengths into micro-batches. We evaluate our approach on real-world instruction datasets using the LLaMA-based model, and our results demonstrate an impressive 86% improvement in inference throughput without compromising effectiveness. Notably, our method is orthogonal to other inference acceleration techniques, making it a valuable addition to many existing toolkits (e.g., FlashAttention, Quantization) for LLM inference.

Recently, long-thought reasoning LLMs, such as OpenAI's O1, adopt extended reasoning processes similar to how humans ponder over complex problems. This reasoning paradigm significantly enhances the model's problem-solving abilities and has achieved promising results. However, long-thought reasoning process leads to a substantial increase in inference time. A pressing challenge is reducing the inference overhead of long-thought LLMs while ensuring accuracy. In this paper, we experimentally demonstrate that long-thought reasoning models struggle to effectively allocate token budgets based on problem difficulty and reasoning redundancies. To address this, we propose Length-Harmonizing Fine-Tuning (O1-Pruner), aiming at minimizing reasoning overhead while maintaining accuracy. This effective fine-tuning method first estimates the LLM's baseline performance through pre-sampling and then uses RL-style fine-tuning to encourage the model to generate shorter reasoning processes under accuracy constraints. This allows the model to achieve efficient reasoning with lower redundancy while maintaining accuracy. Experiments on various mathematical reasoning benchmarks show that O1-Pruner not only significantly reduces inference overhead but also achieves higher accuracy, providing a novel and promising solution to this challenge. Our code is coming soon at https://github.com/StarDewXXX/O1-Pruner

Although Large Vision-Language Models (LVLMs) have achieved impressive results, their high computational cost poses a significant barrier to wider application. To enhance inference efficiency, most existing approaches depend on parameter-dependent or token-dependent strategies to reduce computational demands. However, these methods typically require complex training processes and struggle to consistently select the most relevant tokens. In this paper, we systematically analyze the above challenges and provide a series of valuable insights for inference acceleration. Based on these findings, we propose a novel framework, the Pruning All-Rounder (PAR). Different from previous works, PAR develops a meta-router to adaptively organize pruning flows across both tokens and layers. With a self-supervised learning manner, our method achieves a superior balance between performance and efficiency. Notably, PAR is highly flexible, offering multiple pruning versions to address a range of pruning scenarios. The code for this work will be made publicly available.

The sparsely gated mixture of experts (MoE) architecture sends different inputs to different subnetworks, i.e., experts, through trainable routers. MoE reduces the training computation significantly for large models, but its deployment can be still memory or computation expensive for some downstream tasks. Model pruning is a popular approach to reduce inference computation, but its application in MoE architecture is largely unexplored. To the best of our knowledge, this paper provides the first provably efficient technique for pruning experts in finetuned MoE models. We theoretically prove that prioritizing the pruning of the experts with a smaller change of the routers l2 norm from the pretrained model guarantees the preservation of test accuracy, while significantly reducing the model size and the computational requirements. Although our theoretical analysis is centered on binary classification tasks on simplified MoE architecture, our expert pruning method is verified on large vision MoE models such as VMoE and E3MoE finetuned on benchmark datasets such as CIFAR10, CIFAR100, and ImageNet.

With the ubiquitous use of modern large language models (LLMs) across industries, the inference serving for these models is ever expanding. Given the high compute and memory requirements of modern LLMs, more and more top-of-the-line GPUs are being deployed to serve these models. Energy availability has come to the forefront as the biggest challenge for data center expansion to serve these models. In this paper, we present the trade-offs brought up by making energy efficiency the primary goal of LLM serving under performance SLOs. We show that depending on the inputs, the model, and the service-level agreements, there are several knobs available to the LLM inference provider to use for being energy efficient. We characterize the impact of these knobs on the latency, throughput, as well as the energy. By exploring these trade-offs, we offer valuable insights into optimizing energy usage without compromising on performance, thereby paving the way for sustainable and cost-effective LLM deployment in data center environments.

Large language models (LLMs) are central to modern natural language processing, delivering exceptional performance in various tasks. However, their intensive computational and memory requirements present challenges, especially for devices with limited DRAM capacity. This paper tackles the challenge of efficiently running LLMs that exceed the available DRAM capacity by storing the model parameters on flash memory but bringing them on demand to DRAM. Our method involves constructing an inference cost model that harmonizes with the flash memory behavior, guiding us to optimize in two critical areas: reducing the volume of data transferred from flash and reading data in larger, more contiguous chunks. Within this flash memory-informed framework, we introduce two principal techniques. First, "windowing'" strategically reduces data transfer by reusing previously activated neurons, and second, "row-column bundling", tailored to the sequential data access strengths of flash memory, increases the size of data chunks read from flash memory. These methods collectively enable running models up to twice the size of the available DRAM, with a 4-5x and 20-25x increase in inference speed compared to naive loading approaches in CPU and GPU, respectively. Our integration of sparsity awareness, context-adaptive loading, and a hardware-oriented design paves the way for effective inference of LLMs on devices with limited memory.