Daily Papers

We present Deep Researcher Agent, an open-source framework that enables large language model (LLM) agents to autonomously conduct deep learning experiments around the clock. Unlike existing AI research assistants that focus on paper writing or code generation, our system addresses the full experiment lifecycle: hypothesis formation, code implementation, training execution, result analysis, and iterative refinement. The framework introduces three key innovations: (1) Zero-Cost Monitoring -- a monitoring paradigm that incurs zero LLM API costs during model training by relying solely on process-level checks and log file reads; (2) Two-Tier Constant-Size Memory -- a memory architecture capped at sim5K characters regardless of runtime duration, preventing the unbounded context growth that plagues long-running agents; and (3) Minimal-Toolset Leader-Worker Architecture -- a multi-agent design where each worker agent is equipped with only 3--5 tools, reducing per-call token overhead by up to 73\%. In sustained deployments spanning 30+ days, the framework autonomously completed 500+ experiment cycles across four concurrent research projects, achieving a 52\% improvement over baseline metrics in one project through 200+ automated experiments -- all at an average LLM cost of \$0.08 per 24-hour cycle. Code is available at https://github.com/Xiangyue-Zhang/auto-deep-researcher-24x7.

Monolithic large language models (LLMs) like GPT-4 have paved the way for modern generative AI applications. Training, serving, and maintaining monolithic LLMs at scale, however, remains prohibitively expensive and challenging. The disproportionate increase in compute-to-memory ratio of modern AI accelerators have created a memory wall, necessitating new methods to deploy AI. Composition of Experts (CoE) is an alternative modular approach that lowers the cost and complexity of training and serving. However, this approach presents two key challenges when using conventional hardware: (1) without fused operations, smaller models have lower operational intensity, which makes high utilization more challenging to achieve; and (2) hosting a large number of models can be either prohibitively expensive or slow when dynamically switching between them. In this paper, we describe how combining CoE, streaming dataflow, and a three-tier memory system scales the AI memory wall. We describe Samba-CoE, a CoE system with 150 experts and a trillion total parameters. We deploy Samba-CoE on the SambaNova SN40L Reconfigurable Dataflow Unit (RDU) - a commercial dataflow accelerator architecture that has been co-designed for enterprise inference and training applications. The chip introduces a new three-tier memory system with on-chip distributed SRAM, on-package HBM, and off-package DDR DRAM. A dedicated inter-RDU network enables scaling up and out over multiple sockets. We demonstrate speedups ranging from 2x to 13x on various benchmarks running on eight RDU sockets compared with an unfused baseline. We show that for CoE inference deployments, the 8-socket RDU Node reduces machine footprint by up to 19x, speeds up model switching time by 15x to 31x, and achieves an overall speedup of 3.7x over a DGX H100 and 6.6x over a DGX A100.

Memory is increasingly central to Large Language Model (LLM) agents operating beyond a single context window, yet most existing systems rely on offline, query-agnostic memory construction that can be inefficient and may discard query-critical information. Although runtime memory utilization is a natural alternative, prior work often incurs substantial overhead and offers limited explicit control over the performance-cost trade-off. In this work, we present BudgetMem, a runtime agent memory framework for explicit, query-aware performance-cost control. BudgetMem structures memory processing as a set of memory modules, each offered in three budget tiers (i.e., Low/Mid/High). A lightweight router performs budget-tier routing across modules to balance task performance and memory construction cost, which is implemented as a compact neural policy trained with reinforcement learning. Using BudgetMem as a unified testbed, we study three complementary strategies for realizing budget tiers: implementation (method complexity), reasoning (inference behavior), and capacity (module model size). Across LoCoMo, LongMemEval, and HotpotQA, BudgetMem surpasses strong baselines when performance is prioritized (i.e., high-budget setting), and delivers better accuracy-cost frontiers under tighter budgets. Moreover, our analysis disentangles the strengths and weaknesses of different tiering strategies, clarifying when each axis delivers the most favorable trade-offs under varying budget regimes.

Large language models (LLMs) have revolutionized numerous applications, yet their deployment remains challenged by memory constraints on local devices. While scaling laws have enhanced LLM capabilities, the primary bottleneck has shifted from capability to availability, emphasizing the need for efficient memory management. Traditional compression methods, such as quantization, often require predefined compression ratios and separate compression processes for each setting, complicating deployment in variable memory environments. In this paper, we introduce BitStack, a novel, training-free weight compression approach that enables megabyte-level trade-offs between memory usage and model performance. By leveraging weight decomposition, BitStack can dynamically adjust the model size with minimal transmission between running memory and storage devices. Our approach iteratively decomposes weight matrices while considering the significance of each parameter, resulting in an approximately 1-bit per parameter residual block in each decomposition iteration. These blocks are sorted and stacked in storage as basic transmission units, with different quantities loaded based on current memory availability. Extensive experiments across a wide range of tasks demonstrate that, despite offering fine-grained size control, BitStack consistently matches or surpasses strong quantization baselines, particularly at extreme compression ratios. To the best of our knowledge, this is the first decomposition-based method that effectively bridges the gap to practical compression techniques like quantization. Code is available at https://github.com/xinghaow99/BitStack.

Ternary weight quantization (e.g., BitNet b1.58) offers a promising path to mitigate the memory bandwidth bottleneck in Large Language Model (LLM) inference. However, conventional compute platforms lack native support for ternary-weight arithmetic, often relying on inefficient dequantization. Lookup table (LUT)-based hardware architectures provide an effective alternative by replacing multiplications with conditional additions, but their design space remains largely unexplored. Existing designs rely on heuristic parameter selection, lacking a systematic understanding of the architectural trade-offs. This work addresses this gap by formalizing the design space of ternary LUT-based accelerators and presenting an open-source hardware generator coupled with an analytical cost model, validated against synthesis in TSMC 16nm technology. By spanning the full architectural space, this framework not only enables rapid design space exploration but also establishes a common footing for fair cross-design evaluation, which was previously hindered by inconsistent instantiations across published accelerators. Using this framework, we challenge several assumptions and design choices in recent literature. We demonstrate that the optimal architecture is fundamentally governed by the activation data type: while LUT-based reuse offers significant gains for high-cost arithmetic (e.g., FP16), it yields diminishing returns for small integer types. Furthermore, we show that maximizing core size consistently improves area density compared to highly tiled approaches. Our optimized designs achieve a 2.2x area reduction compared to multiplier-based baselines. Moreover, by benchmarking state-of-the-art implementations against our model, we reveal that correcting suboptimal parameters yields up to a 1.2x area improvement.

While Mixture of Experts (MoE) models achieve remarkable efficiency by activating only subsets of parameters, they suffer from high memory access costs during inference. Memory-layer architectures offer an appealing alternative with very few memory access, but previous attempts like UltraMem have only matched the performance of 2-expert MoE models, falling significantly short of state-of-the-art 8-expert configurations. We present UltraMemV2, a redesigned memory-layer architecture that closes this performance gap. Our approach introduces five key improvements: integrating memory layers into every transformer block, simplifying value expansion with single linear projections, adopting FFN-based value processing from PEER, implementing principled parameter initialization, and rebalancing memory-to-FFN computation ratios. Through extensive evaluation, we demonstrate that UltraMemV2 achieves performance parity with 8-expert MoE models under same computation and parameters but significantly low memory access. Notably, UltraMemV2 shows superior performance on memory-intensive tasks, with improvements of +1.6 points on long-context memorization, +6.2 points on multi-round memorization, and +7.9 points on in-context learning. We validate our approach at scale with models up to 2.5B activated parameters from 120B total parameters, and establish that activation density has greater impact on performance than total sparse parameter count. Our work brings memory-layer architectures to performance parity with state-of-the-art MoE models, presenting a compelling alternative for efficient sparse computation.

Large Language models (LLMs) have become a research hotspot. To accelerate the inference of LLMs, storing computed caches in memory has become the standard technique. However, as the inference length increases, growing KV caches might lead to out-of-memory issues. Many existing methods address this issue through KV cache compression, primarily by preserving key tokens throughout all layers to reduce information loss. Most of them allocate a uniform budget size for each layer to retain. However, we observe that the minimum budget sizes needed to retain essential information vary across layers and models based on the perspectives of attention and hidden state output. Building on this observation, this paper proposes a simple yet effective KV cache compression method that leverages layer uncertainty to allocate budget size for each layer. Experimental results show that the proposed method can reduce memory usage of the KV caches to only sim20\% when compared to Full KV inference while achieving nearly lossless performance.

Memory layers use a trainable key-value lookup mechanism to add extra parameters to a model without increasing FLOPs. Conceptually, sparsely activated memory layers complement compute-heavy dense feed-forward layers, providing dedicated capacity to store and retrieve information cheaply. This work takes memory layers beyond proof-of-concept, proving their utility at contemporary scale. On downstream tasks, language models augmented with our improved memory layer outperform dense models with more than twice the computation budget, as well as mixture-of-expert models when matched for both compute and parameters. We find gains are especially pronounced for factual tasks. We provide a fully parallelizable memory layer implementation, demonstrating scaling laws with up to 128B memory parameters, pretrained to 1 trillion tokens, comparing to base models with up to 8B parameters.

Constant bit-size Transformers are known to be Turing complete, but existing constructions require Ω(s(n)) chain-of-thought (CoT) steps per simulated Turing machine (TM) step, leading to impractical reasoning lengths. In this paper, we significantly reduce this efficiency gap by proving that any (t(n),s(n))-bounded multi-tape TM can be simulated by a constant bit-size Transformer with an optimal O(s(n))-long context window and only O(s(n)^c) CoT steps per TM step, where c>0 can be made arbitrarily small by letting the Transformers' head-layer product sufficiently large. In addition, our construction shows that sparse attention with fixed geometric offsets suffices for efficient universal computation. Our proof leverages multi-queue TMs as a bridge. The main technical novelty is a more efficient simulation of multi-tape TMs by synchronous multi-queue TMs, improving both time and space complexity under stricter model assumptions.

The deployment of Large Language Models (LLMs) on resource-constrained edge devices is increasingly hindered by prohibitive memory and computational requirements. While ternary quantization offers a compelling solution by reducing weights to {-1, 0, +1}, current implementations suffer from a fundamental misalignment with commodity hardware. Most existing methods must choose between 2-bit aligned packing, which incurs significant bit wastage, or 1.67-bit irregular packing, which degrades inference speed. To resolve this tension, we propose Sherry, a hardware-efficient ternary quantization framework. Sherry introduces a 3:4 fine-grained sparsity that achieves a regularized 1.25-bit width by packing blocks of four weights into five bits, restoring power-of-two alignment. Furthermore, we identify weight trapping issue in sparse ternary training, which leads to representational collapse. To address this, Sherry introduces Arenas, an annealing residual synapse mechanism that maintains representational diversity during training. Empirical evaluations on LLaMA-3.2 across five benchmarks demonstrate that Sherry matches state-of-the-art ternary performance while significantly reducing model size. Notably, on an Intel i7-14700HX CPU, our 1B model achieves zero accuracy loss compared to SOTA baselines while providing 25% bit savings and 10% speed up. The code is available at https://github.com/Tencent/AngelSlim .

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.

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.

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.

Agentic large language model (LLM) systems rely on external memory for long-horizon state and concurrent multi-agent execution, but centralized indexes and heuristic partitions become bottlenecks as memory volume and parallel access grow. We present ShardMemo, a budgeted tiered memory service with Tier A per-agent working state, Tier B sharded evidence with shard-local approximate nearest neighbor (ANN) indexes, and Tier C, a versioned skill library. Tier B enforces scope-before-routing: structured eligibility constraints mask ineligible shards before routing or ANN search. We cast shard probing as masked mixture-of-experts (MoE) routing over eligible shards, probing up to B_{probe} shards via Top-B_{probe} or adaptive Top-P, and use cost-aware gating over profile/observation/session shard families; the router is trained from evidence-to-shard supervision. On LoCoMo, ShardMemo improves over the strongest baseline (GAM) by +5.11 to +6.82 F1 across question categories. Under a fixed-budget routing setting (B_{probe}=3), ShardMemo improves over cosine-to-prototype shard routing by +6.87 F1 while reducing retrieval work (VecScan 521->414, -20.5%) and p95 latency (95->76 ms). On long-context HotpotQA, ShardMemo achieves 63.41/61.88/57.95 F1 at 56K/224K/448K tokens. On ToolBench, Tier C reaches 0.97 Precision@3 and 1.94 StepRed (+10.2% and +7.2% over embedding-similarity retrieval).

The optimal bit-width for achieving the best trade-off between quantized model size and accuracy has been a subject of ongoing debate. While some advocate for 4-bit quantization, others propose that 1.58-bit offers superior results. However, the lack of a cohesive framework for different bits has left such conclusions relatively tenuous. We present ParetoQ, the first unified framework that facilitates rigorous comparisons across 1-bit, 1.58-bit, 2-bit, 3-bit, and 4-bit quantization settings. Our findings reveal a notable learning transition between 2 and 3 bits: For 3-bits and above, the fine-tuned models stay close to their original pre-trained distributions, whereas for learning 2-bit networks or below, the representations change drastically. By optimizing training schemes and refining quantization functions, ParetoQ surpasses all previous methods tailored to specific bit widths. Remarkably, our ParetoQ ternary 600M-parameter model even outperforms the previous SoTA ternary 3B-parameter model in accuracy, using only one-fifth of the parameters. Extensive experimentation shows that ternary, 2-bit, and 3-bit quantization maintains comparable performance in the size-accuracy trade-off and generally exceeds 4-bit and binary quantization. Considering hardware constraints, 2-bit quantization offers promising potential for memory reduction and speedup.

Existing works increasingly adopt memory-centric mechanisms to process long contexts in a segment manner, and effective memory management is one of the key capabilities that enables large language models to effectively propagate information across the entire sequence. Therefore, leveraging reward models (RMs) to automatically and reliably evaluate memory quality is critical. In this work, we introduce MemoryRewardBench, the first benchmark to systematically study the ability of RMs to evaluate long-term memory management processes. MemoryRewardBench covers both long-context comprehension and long-form generation tasks, featuring 10 distinct settings with different memory management patterns, with context length ranging from 8K to 128K tokens. Evaluations on 13 cutting-edge RMs indicate a diminishing performance gap between open-source and proprietary models, with newer-generation models consistently outperforming their predecessors regardless of parameter count. We further expose the capabilities and fundamental limitations of current RMs in evaluating LLM memory management across diverse settings.

We study how to endow GUI agents with scalable memory that help generalize across unfamiliar interfaces and long-horizon tasks. Prior GUI agents compress past trajectories into text tokens, which balloons context length and misses decisive visual cues (e.g., exact widget size and position). We propose a continuous memory that encodes each GUI trajectory into a fixed-length sequence of continuous embeddings using the VLM itself as an encoder; these embeddings are plugged directly into the backbone's input layer, sharply reducing context cost while preserving fine-grained visual information. As memory size and retrieval depth increase, performance improves monotonically, unlike text memories that degrade with long prompts. To grow memory at low cost, we introduce an auto-scaling data flywheel that (i) discovers new environments via search, (ii) synthesizes tasks with an open-source VLM, (iii) rolls out trajectories with the agent, and (iv) verifies success with the same VLM. Using this pipeline, we collect 100k+ trajectories for about \$4000 and fine-tune only the memory encoder (LoRA on a Q-Former, 1.2\% parameters) with 1,500 samples. On real-world GUI benchmarks, our memory-augmented agent consistently improves success rates under long horizons and distribution shifts. Notably, Qwen-2.5-VL-7B + continuous memory achieves performance comparable to state-of-the-art closed-source models (e.g., GPT-4o, Claude-4).

The impressive performance gains of modern language models currently rely on scaling parameters: larger models store more world knowledge and reason better. Yet compressing all world knowledge into parameters is unnecessary, as only a fraction is used per prompt, and impractical for edge devices with limited inference-time memory and compute. We address this shortcoming by a memory-augmented architecture and a pretraining strategy aligned with existing hardware paradigms. We introduce small language models that access large hierarchical parametric memory banks encoding world knowledge. During pretraining and inference, we fetch a small, context-dependent memory block and add it to the model. Our pretraining learns to store long-tail world knowledge in the memory parameters, while the small language model acts as an anchor capturing common knowledge and general reasoning abilities. Through trillion-token-scale experiments, we show significant gains: a 160M-parameters model augmented with an 18M-parameters memory fetched from a 4.6B memory bank obtains comparable performance to a regular model with more than 2x the parameters. Through extensive experiments, we study the optimal type and size of parametric memories in transformers, scaling them to over 21B parameters. We find that our proposed hierarchical feed-forward memories work robustly across transformer architectures, whether added during pretraining or post-hoc.

Recurrent neural networks (RNNs) with deep test-time memorization modules, such as Titans and TTT, represent a promising, linearly-scaling paradigm distinct from Transformers. While these expressive models do not yet match the peak performance of state-of-the-art Transformers, their potential has been largely untapped due to prohibitively slow training and low hardware utilization. Existing parallelization methods force a fundamental conflict governed by the chunksize hyperparameter: large chunks boost speed but degrade performance, necessitating a fixed, suboptimal compromise. To solve this challenge, we introduce TNT, a novel training paradigm that decouples training efficiency from inference performance through a two-stage process. Stage one is an efficiency-focused pre-training phase utilizing a hierarchical memory. A global module processes large, hardware-friendly chunks for long-range context, while multiple parallel local modules handle fine-grained details. Crucially, by periodically resetting local memory states, we break sequential dependencies to enable massive context parallelization. Stage two is a brief fine-tuning phase where only the local memory modules are adapted to a smaller, high-resolution chunksize, maximizing accuracy with minimal overhead. Evaluated on Titans and TTT models, TNT achieves a substantial acceleration in training speed-up to 17 times faster than the most accurate baseline configuration - while simultaneously improving model accuracy. This improvement removes a critical scalability barrier, establishing a practical foundation for developing expressive RNNs and facilitating future work to close the performance gap with Transformers.

Large deep learning models offer significant accuracy gains, but training billions to trillions of parameters is challenging. Existing solutions such as data and model parallelisms exhibit fundamental limitations to fit these models into limited device memory, while obtaining computation, communication and development efficiency. We develop a novel solution, Zero Redundancy Optimizer (ZeRO), to optimize memory, vastly improving training speed while increasing the model size that can be efficiently trained. ZeRO eliminates memory redundancies in data- and model-parallel training while retaining low communication volume and high computational granularity, allowing us to scale the model size proportional to the number of devices with sustained high efficiency. Our analysis on memory requirements and communication volume demonstrates: ZeRO has the potential to scale beyond 1 Trillion parameters using today's hardware. We implement and evaluate ZeRO: it trains large models of over 100B parameter with super-linear speedup on 400 GPUs, achieving throughput of 15 Petaflops. This represents an 8x increase in model size and 10x increase in achievable performance over state-of-the-art. In terms of usability, ZeRO can train large models of up to 13B parameters (e.g., larger than Megatron GPT 8.3B and T5 11B) without requiring model parallelism which is harder for scientists to apply. Last but not the least, researchers have used the system breakthroughs of ZeRO to create the world's largest language model (Turing-NLG, 17B parameters) with record breaking accuracy.

Transformer-based large language models (LLMs) are constrained by the fixed context window of the underlying transformer architecture, hindering their ability to produce long and coherent outputs. Memory-augmented LLMs are a promising solution, but current approaches cannot handle long output generation tasks since they (1) only focus on reading memory and reduce its evolution to the concatenation of new memories or (2) use very specialized memories that cannot adapt to other domains. This paper presents L2MAC, the first practical LLM-based general-purpose stored-program automatic computer (von Neumann architecture) framework, an LLM-based multi-agent system, for long and consistent output generation. Its memory has two components: the instruction registry, which is populated with a prompt program to solve the user-given task, and a file store, which will contain the final and intermediate outputs. Each instruction in turn is executed by a separate LLM agent, whose context is managed by a control unit capable of precise memory reading and writing to ensure effective interaction with the file store. These components enable L2MAC to generate extensive outputs, bypassing the constraints of the finite context window while producing outputs that fulfill a complex user-specified task. We empirically demonstrate that L2MAC achieves state-of-the-art performance in generating large codebases for system design tasks, significantly outperforming other coding methods in implementing the detailed user-specified task; we show that L2MAC works for general-purpose extensive text-based tasks, such as writing an entire book; and we provide valuable insights into L2MAC's performance improvement over existing methods.

We describe a family of architectures to support transductive inference by allowing memory to grow to a finite but a-priori unknown bound while making efficient use of finite resources for inference. Current architectures use such resources to represent data either eidetically over a finite span ("context" in Transformers), or fading over an infinite span (in State Space Models, or SSMs). Recent hybrid architectures have combined eidetic and fading memory, but with limitations that do not allow the designer or the learning process to seamlessly modulate the two, nor to extend the eidetic memory span. We leverage ideas from Stochastic Realization Theory to develop a class of models called B'MOJO to seamlessly combine eidetic and fading memory within an elementary composable module. The overall architecture can be used to implement models that can access short-term eidetic memory "in-context," permanent structural memory "in-weights," fading memory "in-state," and long-term eidetic memory "in-storage" by natively incorporating retrieval from an asynchronously updated memory. We show that Transformers, existing SSMs such as Mamba, and hybrid architectures such as Jamba are special cases of B'MOJO and describe a basic implementation, to be open sourced, that can be stacked and scaled efficiently in hardware. We test B'MOJO on transductive inference tasks, such as associative recall, where it outperforms existing SSMs and Hybrid models; as a baseline, we test ordinary language modeling where B'MOJO achieves perplexity comparable to similarly-sized Transformers and SSMs up to 1.4B parameters, while being up to 10% faster to train. Finally, we show that B'MOJO's ability to modulate eidetic and fading memory results in better inference on longer sequences tested up to 32K tokens, four-fold the length of the longest sequences seen during training.

We rethink test-time scaling laws from a practical efficiency perspective, revealing that the effectiveness of smaller models is significantly overestimated. Prior work, grounded in compute-optimality, overlooks critical memory access bottlenecks introduced by inference-time strategies (e.g., Best-of-N, long CoTs). Our holistic analysis, spanning models from 0.6B to 32B parameters, reveals a new Kinetics Scaling Law that better guides resource allocation by incorporating both computation and memory access costs. Kinetics Scaling Law suggests that test-time compute is more effective when used on models above a threshold than smaller ones. A key reason is that in TTS, attention, rather than parameter count, emerges as the dominant cost factor. Motivated by this, we propose a new scaling paradigm centered on sparse attention, which lowers per-token cost and enables longer generations and more parallel samples within the same resource budget. Empirically, we show that sparse attention models consistently outperform dense counterparts, achieving over 60 points gains in low-cost regimes and over 5 points gains in high-cost regimes for problem-solving accuracy on AIME, encompassing evaluations on state-of-the-art MoEs. These results suggest that sparse attention is essential for realizing the full potential of test-time scaling because, unlike training, where parameter scaling saturates, test-time accuracy continues to improve through increased generation. The code is available at https://github.com/Infini-AI-Lab/Kinetics.

Efficient inference with Large Language Models (LLMs) increasingly relies on Key-Value (KV) caches to store previously computed key and value vectors at each layer. These caches are essential to minimize redundant computation during autoregressive token generation, lowering computational complexity from quadratic to linear. However, the growth of KV caches has posed significant system-level challenges, particularly as model sizes increase, context lengths grow, and concurrent requests compete for limited memory resources. Even though several recent frameworks for KV cache management have emerged, their comparative trade-offs in memory consumption and inference performance have not been fully understood, especially under varying request sizes and model configurations. In this work, we conduct an empirical study of three state-of-the-art KV cache management frameworks: vLLM, InfiniGen, and H2O. These frameworks employ techniques such as tensor offloading, token eviction heuristics, and speculative scheduling to balance memory usage and performance. We evaluate their performance in terms of a range of metrics such as latency, throughput, and memory usage across a spectrum of key parameters including request rates, model sizes, and sparsity levels. Our results pinpoint the conditions for each framework to perform the best, revealing the most suitable selection and configuration of KV cache strategies under memory and performance constraints.

Large Language Models (LLMs) have achieved remarkable advancements, but their monolithic nature presents challenges in terms of scalability, cost, and customization. This paper introduces the Composition of Experts (CoE), a modular compound AI system leveraging multiple expert LLMs. CoE leverages a router to dynamically select the most appropriate expert for a given input, enabling efficient utilization of resources and improved performance. We formulate the general problem of training a CoE and discuss inherent complexities associated with it. We propose a two-step routing approach to address these complexities that first uses a router to classify the input into distinct categories followed by a category-to-expert mapping to obtain desired experts. CoE offers a flexible and cost-effective solution to build compound AI systems. Our empirical evaluation demonstrates the effectiveness of CoE in achieving superior performance with reduced computational overhead. Given that CoE comprises of many expert LLMs it has unique system requirements for cost-effective serving. We present an efficient implementation of CoE leveraging SambaNova SN40L RDUs unique three-tiered memory architecture. CoEs obtained using open weight LLMs Qwen/Qwen2-7B-Instruct, google/gemma-2-9b-it, google/gemma-2-27b-it, meta-llama/Llama-3.1-70B-Instruct and Qwen/Qwen2-72B-Instruct achieve a score of 59.4 with merely 31 billion average active parameters on Arena-Hard and a score of 9.06 with 54 billion average active parameters on MT-Bench.

Deploying Large Language Models (LLMs) on resource-constrained edge devices faces critical bottlenecks in memory bandwidth and power consumption. While ternary quantization (e.g., BitNet b1.58) significantly reduces model size, its direct deployment on general-purpose hardware is hindered by workload imbalance, bandwidth-bound decoding, and strict data dependencies. To address these challenges, we propose VitaLLM, a hardware-software co-designed accelerator tailored for efficient ternary LLM inference. We introduce a heterogeneous Dual-Core Compute Strategy that synergizes specialized TINT-Cores for massive ternary projections with a unified BoothFlex-Core for mixed-precision attention, ensuring high utilization across both compute-bound prefill and bandwidth-bound decode stages. Furthermore, we develop a Leading One Prediction (LOP) mechanism to prune redundant Key-Value (KV) cache fetches and a Dependency-Aware Scheduling framework to hide the latency of nonlinear operations. Implemented in TSMC 16nm technology, VitaLLM achieves a decoding throughput of 70.70 tokens/s within an ultra-compact area of 0.223 mm^2 and a power consumption of 65.97 mW. The design delivers a superior Figure of Merit (FOM) of 17.4 TOPS/mm^2/W, significantly outperforming state-of-the-art accelerators. Finally, we explore an extended bit-serial design (BoothFlex-BS) to demonstrate the architecture's adaptability for precision-agile inference.

Real-world applications require the classification model to adapt to new classes without forgetting old ones. Correspondingly, Class-Incremental Learning (CIL) aims to train a model with limited memory size to meet this requirement. Typical CIL methods tend to save representative exemplars from former classes to resist forgetting, while recent works find that storing models from history can substantially boost the performance. However, the stored models are not counted into the memory budget, which implicitly results in unfair comparisons. We find that when counting the model size into the total budget and comparing methods with aligned memory size, saving models do not consistently work, especially for the case with limited memory budgets. As a result, we need to holistically evaluate different CIL methods at different memory scales and simultaneously consider accuracy and memory size for measurement. On the other hand, we dive deeply into the construction of the memory buffer for memory efficiency. By analyzing the effect of different layers in the network, we find that shallow and deep layers have different characteristics in CIL. Motivated by this, we propose a simple yet effective baseline, denoted as MEMO for Memory-efficient Expandable MOdel. MEMO extends specialized layers based on the shared generalized representations, efficiently extracting diverse representations with modest cost and maintaining representative exemplars. Extensive experiments on benchmark datasets validate MEMO's competitive performance. Code is available at: https://github.com/wangkiw/ICLR23-MEMO

Long-context LLMs have enabled numerous downstream applications but also introduced significant challenges related to computational and memory efficiency. To address these challenges, optimizations for long-context inference have been developed, centered around the KV cache. However, existing benchmarks often evaluate in single-request, neglecting the full lifecycle of the KV cache in real-world use. This oversight is particularly critical, as KV cache reuse has become widely adopted in LLMs inference frameworks, such as vLLM and SGLang, as well as by LLM providers, including OpenAI, Microsoft, Google, and Anthropic. To address this gap, we introduce SCBench(SharedContextBench), a comprehensive benchmark for evaluating long-context methods from a KV cachecentric perspective: 1) KV cache generation, 2) KV cache compression, 3) KV cache retrieval, 4) KV cache loading. Specifically, SCBench uses test examples with shared context, ranging 12 tasks with two shared context modes, covering four categories of long-context capabilities: string retrieval, semantic retrieval, global information, and multi-task. With it, we provide an extensive KV cache-centric analysis of eight categories long-context solutions, including Gated Linear RNNs, Mamba-Attention hybrids, and efficient methods such as sparse attention, KV cache dropping, quantization, retrieval, loading, and prompt compression. The evaluation is conducted on 8 long-context LLMs. Our findings show that sub-O(n) memory methods suffer in multi-turn scenarios, while sparse encoding with O(n) memory and sub-O(n^2) pre-filling computation perform robustly. Dynamic sparsity yields more expressive KV caches than static patterns, and layer-level sparsity in hybrid architectures reduces memory usage with strong performance. Additionally, we identify attention distribution shift issues in long-generation scenarios. https://aka.ms/SCBench.

Efficient real-world deployments of large language models (LLMs) rely on Key-Value (KV) caching for processing and generating long outputs, reducing the need for repetitive computation. For large contexts, Key-Value caches can take up tens of gigabytes of device memory, as they store vector representations for each token and layer. Recent work has shown that the cached vectors can be compressed through quantization, pruning or merging, but these techniques often compromise quality towards higher compression rates. In this work, we aim to improve Key & Value compression by exploiting two observations: 1) the inherent dependencies between keys and values across different layers, and 2) high-compression mechanisms for internal network states. We propose AQUA-KV, an adaptive quantization for Key-Value caches that relies on compact adapters to exploit existing dependencies between Keys and Values, and aims to "optimally" compress the information that cannot be predicted. AQUA-KV significantly improves compression rates, while maintaining high accuracy on state-of-the-art LLM families. On Llama 3.2 LLMs, we achieve near-lossless inference at 2-2.5 bits per value with under 1% relative error in perplexity and LongBench scores. AQUA-KV is one-shot, simple, and efficient: it can be calibrated on a single GPU within 1-6 hours, even for 70B models.

Retrieval-augmented generation (RAG) has become the default strategy for providing large language model (LLM) agents with contextual knowledge. Yet RAG treats memory as a stateless lookup table: information persists indefinitely, retrieval is read-only, and temporal continuity is absent. We define the Continuum Memory Architecture (CMA), a class of systems that maintain and update internal state across interactions through persistent storage, selective retention, associative routing, temporal chaining, and consolidation into higher-order abstractions. Rather than disclosing implementation specifics, we specify the architectural requirements CMA imposes and show consistent behavioral advantages on tasks that expose RAG's structural inability to accumulate, mutate, or disambiguate memory. The empirical probes (knowledge updates, temporal association, associative recall, contextual disambiguation) demonstrate that CMA is a necessary architectural primitive for long-horizon agents while highlighting open challenges around latency, drift, and interpretability.

High-speed packet processing on multicore CPUs places extreme demands on memory allocators. In systems like DPDK, fixed-size memory pools back packet buffers (mbufs) to avoid costly dynamic allocation. However, even DPDK's optimized mempool faces scalability limits: lock contention on the shared ring, cache-coherence ping-pong between cores, and heavy TLB pressure from thousands of small pages. To mitigate these issues, DPDK typically uses explicit huge pages (2 MB or 1 GB) for its memory pools. This reduces TLB misses but requires manual configuration and can lead to fragmentation and inflexibility. We propose TurboMem, a novel C++ template-based memory pool that addresses these challenges. TurboMem combines a fully lock-free design (using atomic stacks and per-core local caches) with Transparent Huge Page (THP) auto merging. By automatically promoting pools to 2 MB pages via madvise(MADV_HUGEPAGE), TurboMem achieves the benefits of huge pages without manual setup. We also enforce strict NUMA locality and CPU affinity, so each core allocates and frees objects from its local node. Using Intel VTune on a single-socket 100 Gbps testbed, we show that TurboMem boosts packet throughput by up to 28% while reducing TLB misses by 41% compared to a standard DPDK mempool with explicit huge pages. These results demonstrate that THP auto-merging can outperform manually reserved huge pages in low-fragmentation scenarios, and that modern C++ lock-free programming yields practical gains in data-plane software. Note: The performance claims reported in this preliminary version (up to 28% higher throughput and 41% fewer TLB misses) are based on mock benchmarks. Comprehensive real-system evaluations using Intel VTune are currently underway and will be presented in a future revision.

Large Language Models (LLMs) have emerged as foundational infrastructure in the pursuit of Artificial General Intelligence (AGI). Despite their remarkable capabilities in language perception and generation, current LLMs fundamentally lack a unified and structured architecture for handling memory. They primarily rely on parametric memory (knowledge encoded in model weights) and ephemeral activation memory (context-limited runtime states). While emerging methods like Retrieval-Augmented Generation (RAG) incorporate plaintext memory, they lack lifecycle management and multi-modal integration, limiting their capacity for long-term knowledge evolution. To address this, we introduce MemOS, a memory operating system designed for LLMs that, for the first time, elevates memory to a first-class operational resource. It builds unified mechanisms for representation, organization, and governance across three core memory types: parametric, activation, and plaintext. At its core is the MemCube, a standardized memory abstraction that enables tracking, fusion, and migration of heterogeneous memory, while offering structured, traceable access across tasks and contexts. MemOS establishes a memory-centric execution framework with strong controllability, adaptability, and evolvability. It fills a critical gap in current LLM infrastructure and lays the groundwork for continual adaptation, personalized intelligence, and cross-platform coordination in next-generation intelligent systems.

This paper introduces a structured memory which can be easily integrated into a neural network. The memory is very large by design and significantly increases the capacity of the architecture, by up to a billion parameters with a negligible computational overhead. Its design and access pattern is based on product keys, which enable fast and exact nearest neighbor search. The ability to increase the number of parameters while keeping the same computational budget lets the overall system strike a better trade-off between prediction accuracy and computation efficiency both at training and test time. This memory layer allows us to tackle very large scale language modeling tasks. In our experiments we consider a dataset with up to 30 billion words, and we plug our memory layer in a state-of-the-art transformer-based architecture. In particular, we found that a memory augmented model with only 12 layers outperforms a baseline transformer model with 24 layers, while being twice faster at inference time. We release our code for reproducibility purposes.

While Mixture-of-Experts (MoE) scales capacity via conditional computation, Transformers lack a native primitive for knowledge lookup, forcing them to inefficiently simulate retrieval through computation. To address this, we introduce conditional memory as a complementary sparsity axis, instantiated via Engram, a module that modernizes classic N-gram embedding for O(1) lookup. By formulating the Sparsity Allocation problem, we uncover a U-shaped scaling law that optimizes the trade-off between neural computation (MoE) and static memory (Engram). Guided by this law, we scale Engram to 27B parameters, achieving superior performance over a strictly iso-parameter and iso-FLOPs MoE baseline. Most notably, while the memory module is expected to aid knowledge retrieval (e.g., MMLU +3.4; CMMLU +4.0), we observe even larger gains in general reasoning (e.g., BBH +5.0; ARC-Challenge +3.7) and code/math domains~(HumanEval +3.0; MATH +2.4). Mechanistic analyses reveal that Engram relieves the backbone's early layers from static reconstruction, effectively deepening the network for complex reasoning. Furthermore, by delegating local dependencies to lookups, it frees up attention capacity for global context, substantially boosting long-context retrieval (e.g., Multi-Query NIAH: 84.2 to 97.0). Finally, Engram establishes infrastructure-aware efficiency: its deterministic addressing enables runtime prefetching from host memory, incurring negligible overhead. We envision conditional memory as an indispensable modeling primitive for next-generation sparse models.

The linear growth of key-value (KV) cache memory and quadratic computational complexity pose significant bottlenecks for large language models (LLMs) in long-context processing. While existing KV cache optimization methods address these challenges through token pruning or feature merging, they often suffer from irreversible information loss or require costly parameter retraining. We propose ZeroMerge, a dynamic zero-shot compression framework that achieves efficient cache management through three key innovations: (1) Fine-grained memory allocation guided by multi-dimensional token importance metrics at head-level granularity, (2) A residual merging mechanism that preserves critical context through compensated attention scoring, and (3) Parameter-free adaptation compatible with diverse LLM architectures without retraining. Comprehensive evaluations across LLaMA-2 model demonstrate that ZeroMerge maintains full-cache performance at 5\% compression ratios while doubling inference throughput at 40K token lengths. The method effectively balances memory efficiency, generation quality, and deployment flexibility, advancing practical long-context LLM applications. The code is available at https://github.com/SusCom-Lab/ZeroMerge.

Attention for transformers is a critical workload that has recently received significant "attention" as a target for custom acceleration. Yet, while prior work succeeds in reducing attention's memory-bandwidth requirements, it creates load imbalance between attention operators (resulting in severe compute under-utilization) and requires on-chip memory that scales with sequence length (which is expected to grow over time). This paper ameliorates these issues, enabling attention with nearly 100% compute utilization, no off-chip memory traffic bottlenecks, and on-chip buffer size requirements that are independent of sequence length. The main conceptual contribution is to use a recently proposed abstraction -- the cascade of Einsums -- to describe, formalize and taxonomize the space of attention algorithms that appear in the literature. In particular, we show how Einsum cascades can be used to infer non-trivial lower bounds on the number of passes a kernel must take through its input data, which has implications for either required on-chip buffer capacity or memory traffic. We show how this notion can be used to meaningfully divide the space of attention algorithms into several categories and use these categories to inform our design process. Based on the above characterization, we propose FuseMax -- a novel mapping of attention onto a spatial array-style architecture. On attention, in an iso-area comparison, FuseMax achieves an average 6.7times speedup over the prior state-of-the-art FLAT while using 79% of the energy. Similarly, on the full end-to-end transformer inference, FuseMax achieves an average 5.3times speedup over FLAT using 83% of the energy.

Large language model (LLM) agents demonstrate strong performance in short-text contexts but often underperform in extended dialogues due to inefficient memory management. Existing approaches face a fundamental trade-off between efficiency and effectiveness: memory compression risks losing critical details required for complex reasoning, while retaining raw text introduces unnecessary computational overhead for simple queries. The crux lies in the limitations of monolithic memory representations and static retrieval mechanisms, which fail to emulate the flexible and proactive memory scheduling capabilities observed in humans, thus struggling to adapt to diverse problem scenarios. Inspired by the principle of cognitive economy, we propose HyMem, a hybrid memory architecture that enables dynamic on-demand scheduling through multi-granular memory representations. HyMem adopts a dual-granular storage scheme paired with a dynamic two-tier retrieval system: a lightweight module constructs summary-level context for efficient response generation, while an LLM-based deep module is selectively activated only for complex queries, augmented by a reflection mechanism for iterative reasoning refinement. Experiments show that HyMem achieves strong performance on both the LOCOMO and LongMemEval benchmarks, outperforming full-context while reducing computational cost by 92.6\%, establishing a state-of-the-art balance between efficiency and performance in long-term memory management.

Agentic memory systems enable large language model (LLM) agents to maintain state across long interactions, supporting long-horizon reasoning and personalization beyond fixed context windows. Despite rapid architectural development, the empirical foundations of these systems remain fragile: existing benchmarks are often underscaled, evaluation metrics are misaligned with semantic utility, performance varies significantly across backbone models, and system-level costs are frequently overlooked. This survey presents a structured analysis of agentic memory from both architectural and system perspectives. We first introduce a concise taxonomy of MAG systems based on four memory structures. Then, we analyze key pain points limiting current systems, including benchmark saturation effects, metric validity and judge sensitivity, backbone-dependent accuracy, and the latency and throughput overhead introduced by memory maintenance. By connecting the memory structure to empirical limitations, this survey clarifies why current agentic memory systems often underperform their theoretical promise and outlines directions for more reliable evaluation and scalable system design.

Self-attention is at the heart of the popular Transformer architecture, yet suffers from quadratic time and memory complexity. The breakthrough FlashAttention algorithm revealed I/O complexity as the true bottleneck in scaling Transformers. Given two levels of memory hierarchy, a fast cache (e.g. GPU on-chip SRAM) and a slow memory (e.g. GPU high-bandwidth memory), the I/O complexity measures the number of accesses to memory. FlashAttention computes attention using N^2d^2{M} I/O operations where N is the dimension of the attention matrix, d the head-dimension and M the cache size. However, is this I/O complexity optimal? The known lower bound only rules out an I/O complexity of o(Nd) when M=Theta(Nd), since the output that needs to be written to slow memory is Omega(Nd). This leads to the main question of our work: Is FlashAttention I/O optimal for all values of M? We resolve the above question in its full generality by showing an I/O complexity lower bound that matches the upper bound provided by FlashAttention for any values of M geq d^2 within any constant factors. Further, we give a better algorithm with lower I/O complexity for M < d^2, and show that it is optimal as well. Moreover, our lower bounds do not rely on using combinatorial matrix multiplication for computing the attention matrix. We show even if one uses fast matrix multiplication, the above I/O complexity bounds cannot be improved. We do so by introducing a new communication complexity protocol for matrix compression, and connecting communication complexity to I/O complexity. To the best of our knowledge, this is the first work to establish a connection between communication complexity and I/O complexity, and we believe this connection could be of independent interest and will find many more applications in proving I/O complexity lower bounds in the future.

While inference-time scaling enables LLMs to carry out increasingly long and capable reasoning traces, the patterns and insights uncovered during these traces are immediately discarded once the context window is reset for a new query. External memory is a natural way to persist these discoveries, and recent work has shown clear benefits for reasoning-intensive tasks. We see an opportunity to make such memories more broadly reusable and scalable by moving beyond instance-based memory entries (e.g. exact query/response pairs, or summaries tightly coupled with the original problem context) toward concept-level memory: reusable, modular abstractions distilled from solution traces and stored in natural language. For future queries, relevant concepts are selectively retrieved and integrated into the prompt, enabling test-time continual learning without weight updates. Our design introduces new strategies for abstracting takeaways from rollouts and retrieving entries for new queries, promoting reuse and allowing memory to expand with additional experiences. We evaluate on ARC-AGI, a benchmark that stresses compositional generalization and abstract reasoning, making it a natural fit for concept memory. Our method yields a 7.5% relative gain over a strong no-memory baseline with performance continuing to scale with inference compute. We find abstract concepts to be the most consistent memory design, outscoring the baseline at all tested inference compute scales. Moreover, dynamically updating memory during test-time outperforms fixed settings, supporting the hypothesis that accumulating and abstracting patterns enables further solutions in a form of self-improvement. Code is available at https://github.com/matt-seb-ho/arc_memo.

Recently, considerable efforts have been directed towards compressing Large Language Models (LLMs), which showcase groundbreaking capabilities across diverse applications but entail significant deployment costs due to their large sizes. Meanwhile, much less attention has been given to mitigating the costs associated with deploying multiple LLMs of varying sizes despite its practical significance. Thus, this paper introduces any-precision LLM, extending the concept of any-precision DNN to LLMs. Addressing challenges in any-precision LLM, we propose a lightweight method for any-precision quantization of LLMs, leveraging a post-training quantization framework, and develop a specialized software engine for its efficient serving. As a result, our solution significantly reduces the high costs of deploying multiple, different-sized LLMs by overlaying LLMs quantized to varying bit-widths, such as 3, 4, ..., n bits, into a memory footprint comparable to a single n-bit LLM. All the supported LLMs with varying bit-widths demonstrate state-of-the-art model quality and inference throughput, proving itself to be a compelling option for deployment of multiple, different-sized LLMs. The source code will be publicly available soon.

Serving LLMs requires substantial memory due to the storage requirements of Key-Value (KV) embeddings in the KV cache, which grows with sequence length. An effective approach to compress KV cache is quantization. However, traditional quantization methods face significant memory overhead due to the need to store quantization constants (at least a zero point and a scale) in full precision per data block. Depending on the block size, this overhead can add 1 or 2 bits per quantized number. We introduce QJL, a new quantization approach that consists of a Johnson-Lindenstrauss (JL) transform followed by sign-bit quantization. In contrast to existing methods, QJL eliminates memory overheads by removing the need for storing quantization constants. We propose an asymmetric estimator for the inner product of two vectors and demonstrate that applying QJL to one vector and a standard JL transform without quantization to the other provides an unbiased estimator with minimal distortion. We have developed an efficient implementation of the QJL sketch and its corresponding inner product estimator, incorporating a lightweight CUDA kernel for optimized computation. When applied across various LLMs and NLP tasks to quantize the KV cache to only 3 bits, QJL demonstrates a more than fivefold reduction in KV cache memory usage without compromising accuracy, all while achieving faster runtime. Codes are available at https://github.com/amirzandieh/QJL.

The deployment of Large Language Models (LLMs) on edge devices is fundamentally constrained by the "Memory Wall" -- a hardware limitation where memory bandwidth, not compute, becomes the bottleneck. Recent 1.58-bit quantization techniques (e.g., BitNet b1.58) dramatically reduce memory footprint but typically incur a perplexity degradation of 20-25% compared to FP16 baselines. In this work, we introduce Hybrid Gated Flow (HGF), a dual-stream architecture that couples a 1.58-bit ternary backbone with a learnable, low-rank FP16 correction path controlled by adaptive gates. Through extensive experiments on the TinyStories dataset across two training regimes (2500 and 3500 steps), we demonstrate that HGF 5.4 achieves a validation loss of 0.9306 compared to BitNet's 1.0294, recovering approximately 55% of the quality gap between pure ternary quantization and the FP16 baseline (0.8490). This recovery is achieved with only ~12-15% memory overhead beyond the ternary backbone. Furthermore, we provide empirical evidence for an emergent phenomenon: quantization as structural regularization. While a full-precision differential attention baseline (Diff_Only) exhibited training instability with validation loss exceeding 1.68, the ternary-anchored HGF maintained robust convergence throughout training. Finally, we report preliminary results extending this architecture to 1.2B and 3B parameter models trained on SlimPajama and FineWeb-Edu. These larger-scale experiments confirm that the architectural stability and quality recovery observed in small-scale proxies scale linearly to production-grade language modeling regimes.

The deployment of Large Language Models (LLMs) on edge devices is fundamentally constrained by the "Memory Wall" the bottleneck where data movement latency outstrips arithmetic throughput. Standard inference runtimes often incur significant overhead through high-level abstractions, dynamic dispatch, and unaligned memory access patterns. In this work, we present a novel "Virtual Tensor Core" architecture implemented in software, optimized specifically for ARM64 microarchitectures (Apple Silicon). By bypassing standard library containers in favor of direct memory mapping (mmap) and implementing hand-tuned NEON SIMD kernels, we achieve a form of "Software-Defined Direct Memory Access (DMA)." Our proposed Tensor Virtualization Layout (TVL) guarantees 100% cache line utilization for weight matrices, while our zero-copy loader eliminates initialization latency. Experimental results on a 110M parameter model demonstrate a stable throughput of >60 tokens/second on M2 hardware. While proprietary hardware accelerators (e.g., Apple AMX) can achieve higher peak throughput, our architecture provides a fully open, portable, and deterministic reference implementation for studying the memory bottleneck on general-purpose ARM silicon, meeting the 200ms psycholinguistic latency threshold without opaque dependencies.

Large Language Models (LLMs) have shown an impressive capability in code generation and, specifically, to automatically implement requirements described in natural language. The LLM effectiveness generally increases with its size: The higher the number of LLM's trainable parameters the better its ability to implement code. However, when it comes to deploying LLM-based code generators, larger LLMs pose significant challenges related to their memory (and, consequently, carbon) footprint. A previous work by Wei et al. proposed to leverage quantization techniques to reduce the memory footprint of LLM-based code generators without substantially degrading their effectiveness. In short, they studied LLMs featuring up to 16B parameters, quantizing their precision from floating point 32 bits down to int 8 bits and showing their limited impact on code generation performance. Given the fast pace at which LLM capabilities and quantization techniques are evolving, in this work we present a differentiated replication of the work by Wei et al. in which we consider (i) on the one side, more recent and larger code-related LLMs, of up to 34B parameters; (ii) the latest advancements in model quantization techniques, which allow pushing the compression to the extreme quantization level of 2 bits per model parameter and; (iii) different types of calibration datasets to guide the quantization process, including code-specific ones. Our empirical evaluation reveals that the new frontier for LLM quantization is 4-bit precision, resulting in an average memory footprint reduction of 70% compared to the original model without observing any significant decrease in performance. Additionally, when the quantization becomes even more extreme (3 and 2 bits), a code-specific calibration dataset helps to limit the loss of performance.

Efficient use of GPU memory is essential for high throughput LLM inference. Prior systems reserved memory for the KV-cache ahead-of-time, resulting in wasted capacity due to internal fragmentation. Inspired by OS-based virtual memory systems, vLLM proposed PagedAttention to enable dynamic memory allocation for KV-cache. This approach eliminates fragmentation, enabling high-throughput LLM serving with larger batch sizes. However, to be able to allocate physical memory dynamically, PagedAttention changes the layout of KV-cache from contiguous virtual memory to non-contiguous virtual memory. This change requires attention kernels to be rewritten to support paging, and serving framework to implement a memory manager. Thus, the PagedAttention model leads to software complexity, portability issues, redundancy and inefficiency. In this paper, we propose vAttention for dynamic KV-cache memory management. In contrast to PagedAttention, vAttention retains KV-cache in contiguous virtual memory and leverages low-level system support for demand paging, that already exists, to enable on-demand physical memory allocation. Thus, vAttention unburdens the attention kernel developer from having to explicitly support paging and avoids re-implementation of memory management in the serving framework. We show that vAttention enables seamless dynamic memory management for unchanged implementations of various attention kernels. vAttention also generates tokens up to 1.97x faster than vLLM, while processing input prompts up to 3.92x and 1.45x faster than the PagedAttention variants of FlashAttention and FlashInfer.

Long-term memory is a cornerstone of human intelligence. Enabling AI to process lifetime-scale information remains a long-standing pursuit in the field. Due to the constraints of full-attention architectures, the effective context length of large language models (LLMs) is typically limited to 1M tokens. Existing approaches, such as hybrid linear attention, fixed-size memory states (e.g., RNNs), and external storage methods like RAG or agent systems, attempt to extend this limit. However, they often suffer from severe precision degradation and rapidly increasing latency as context length grows, an inability to dynamically modify memory content, or a lack of end-to-end optimization. These bottlenecks impede complex scenarios like large-corpus summarization, Digital Twins, and long-history agent reasoning, while limiting memory capacity and slowing inference. We present Memory Sparse Attention (MSA), an end-to-end trainable, efficient, and massively scalable memory model framework. Through core innovations including scalable sparse attention and document-wise RoPE, MSA achieves linear complexity in both training and inference while maintaining exceptional stability, exhibiting less than 9% degradation when scaling from 16K to 100M tokens. Furthermore, KV cache compression, combined with Memory Parallel, enables 100M-token inference on 2xA800 GPUs. We also propose Memory Interleaving to facilitate complex multi-hop reasoning across scattered memory segments. MSA significantly surpasses frontier LLMs, state-of-the-art RAG systems, and leading memory agents in long-context benchmarks. These results demonstrate that by decoupling memory capacity from reasoning, MSA provides a scalable foundation to endow general-purpose models with intrinsic, lifetime-scale memory.

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 Language Models (LLMs) represent a revolutionary advancement in the contemporary landscape of artificial general intelligence (AGI). As exemplified by ChatGPT, LLM-based applications necessitate minimal response latency and maximal throughput for inference serving. However, due to the unpredictability of LLM execution, the first-come-first-serve (FCFS) scheduling policy employed by current LLM serving systems suffers from head-of-line (HoL) blocking issues and long job response times. In this paper, we propose a new efficient LLM inference serving framework, named ALISE. The key design paradigm of ALISE is to leverage a novel speculative scheduler by estimating the execution time for each job and exploiting such prior knowledge to assign appropriate job priority orders, thus minimizing potential queuing delays for heterogeneous workloads. Furthermore, to mitigate the memory overhead of the intermediate key-value (KV) cache, we employ a priority-based adaptive memory management protocol and quantization-based compression techniques. Evaluations demonstrate that in comparison to the state-of-the-art solution vLLM, ALISE improves the throughput of inference serving by up to 1.8x and 2.1x under the same latency constraint on the Alpaca and ShareGPT datasets, respectively.

The advent of 1-bit large language models (LLMs), led by BitNet b1.58, has spurred interest in ternary LLMs. Despite this, research and practical applications focusing on efficient edge inference for ternary LLMs remain scarce. To bridge this gap, we introduce Bitnet.cpp, an inference system optimized for BitNet b1.58 and ternary LLMs. Given that mixed-precision matrix multiplication (mpGEMM) constitutes the bulk of inference time in ternary LLMs, Bitnet.cpp incorporates a novel mpGEMM library to facilitate sub-2-bits-per-weight, efficient and lossless inference. The library features two core solutions: Ternary Lookup Table (TL), which addresses spatial inefficiencies of previous bit-wise methods, and Int2 with a Scale (I2_S), which ensures lossless edge inference, both enabling high-speed inference. Our experiments show that Bitnet.cpp achieves up to a 6.25x increase in speed over full-precision baselines and up to 2.32x over low-bit baselines, setting new benchmarks in the field. Additionally, we expand TL to element-wise lookup table (ELUT) for low-bit LLMs in the appendix, presenting both theoretical and empirical evidence of its considerable potential. Bitnet.cpp is publicly available at https://github.com/microsoft/BitNet/tree/paper , offering a sophisticated solution for the efficient and practical deployment of edge LLMs.

Extensive system scales (i.e. thousands of GPU/TPUs) and prolonged training periods (i.e. months of pretraining) significantly escalate the probability of failures when training large language models (LLMs). Thus, efficient and reliable fault-tolerance methods are in urgent need. Checkpointing is the primary fault-tolerance method to periodically save parameter snapshots from GPU memory to disks via CPU memory. In this paper, we identify the frequency of existing checkpoint-based fault-tolerance being significantly limited by the storage I/O overheads, which results in hefty re-training costs on restarting from the nearest checkpoint. In response to this gap, we introduce an in-memory fault-tolerance framework for large-scale LLM pretraining. The framework boosts the efficiency and reliability of fault tolerance from three aspects: (1) Reduced Data Transfer and I/O: By asynchronously caching parameters, i.e., sharded model parameters, optimizer states, and RNG states, to CPU volatile memory, Our framework significantly reduces communication costs and bypasses checkpoint I/O. (2) Enhanced System Reliability: Our framework enhances parameter protection with a two-layer hierarchy: snapshot management processes (SMPs) safeguard against software failures, together with Erasure Coding (EC) protecting against node failures. This double-layered protection greatly improves the survival probability of the parameters compared to existing checkpointing methods. (3) Improved Snapshotting Frequency: Our framework achieves more frequent snapshotting compared with asynchronous checkpointing optimizations under the same saving time budget, which improves the fault tolerance efficiency. Empirical results demonstrate that Our framework minimizes the overhead of fault tolerance of LLM pretraining by effectively leveraging redundant CPU resources.

With the emergence of pre-trained vision-language models like CLIP, how to adapt them to various downstream classification tasks has garnered significant attention in recent research. The adaptation strategies can be typically categorized into three paradigms: zero-shot adaptation, few-shot adaptation, and the recently-proposed training-free few-shot adaptation. Most existing approaches are tailored for a specific setting and can only cater to one or two of these paradigms. In this paper, we introduce a versatile adaptation approach that can effectively work under all three settings. Specifically, we propose the dual memory networks that comprise dynamic and static memory components. The static memory caches training data knowledge, enabling training-free few-shot adaptation, while the dynamic memory preserves historical test features online during the testing process, allowing for the exploration of additional data insights beyond the training set. This novel capability enhances model performance in the few-shot setting and enables model usability in the absence of training data. The two memory networks employ the same flexible memory interactive strategy, which can operate in a training-free mode and can be further enhanced by incorporating learnable projection layers. Our approach is tested across 11 datasets under the three task settings. Remarkably, in the zero-shot scenario, it outperforms existing methods by over 3\% and even shows superior results against methods utilizing external training data. Additionally, our method exhibits robust performance against natural distribution shifts. Codes are available at https://github.com/YBZh/DMN.

Modern Large Language Models (LLMs) are increasingly trained to support very large context windows. We present Compactor, a training-free, query-agnostic KV compression strategy that uses approximate leverage scores to determine token importance. We show that Compactor can achieve the same performance as competing methods while retaining 20% fewer tokens in both synthetic and real-world context tasks, while being more task-robust. We further introduce a procedure for context-calibrated compression: inferring the maximum compression a given context supports before significant performance loss. Using context-calibrated compression, we show that Compactor achieves full KV performance on Longbench while reducing the KV memory burden by 68%, on average. To demonstrate the efficacy and generalizability of our approach, we apply Compactor to 27 synthetic and real-world tasks from RULER and Longbench, with models from both the Qwen 2.5 and Llama 3.1 families. Finally, we release compactor-vllm, an inference engine and suite of optimized Triton kernels designed to efficiently support the sparse, non-contiguous memory access patterns inherent to compressed KV caches. This work demonstrates that Compactor offers a practical, high-performance solution for alleviating the memory bottleneck in modern LLM deployment.

Instructing the model to generate a sequence of intermediate steps, a.k.a., a chain of thought (CoT), is a highly effective method to improve the accuracy of large language models (LLMs) on arithmetics and symbolic reasoning tasks. However, the mechanism behind CoT remains unclear. This work provides a theoretical understanding of the power of CoT for decoder-only transformers through the lens of expressiveness. Conceptually, CoT empowers the model with the ability to perform inherently serial computation, which is otherwise lacking in transformers, especially when depth is low. Given input length n, previous works have shown that constant-depth transformers with finite precision poly(n) embedding size can only solve problems in TC^0 without CoT. We first show an even tighter expressiveness upper bound for constant-depth transformers with constant-bit precision, which can only solve problems in AC^0, a proper subset of TC^0. However, with T steps of CoT, constant-depth transformers using constant-bit precision and O(log n) embedding size can solve any problem solvable by boolean circuits of size T. Empirically, enabling CoT dramatically improves the accuracy for tasks that are hard for parallel computation, including the composition of permutation groups, iterated squaring, and circuit value problems, especially for low-depth transformers.

The widespread of Large Language Models (LLMs) marks a significant milestone in generative AI. Nevertheless, the increasing context length and batch size in offline LLM inference escalate the memory requirement of the key-value (KV) cache, which imposes a huge burden on the GPU VRAM, especially for resource-constraint scenarios (e.g., edge computing and personal devices). Several cost-effective solutions leverage host memory or SSDs to reduce storage costs for offline inference scenarios and improve the throughput. Nevertheless, they suffer from significant performance penalties imposed by intensive KV cache accesses due to limited PCIe bandwidth. To address these issues, we propose InstInfer, a novel LLM inference system that offloads the most performance-critical computation (i.e., attention in decoding phase) and data (i.e., KV cache) parts to Computational Storage Drives (CSDs), which minimize the enormous KV transfer overheads. InstInfer designs a dedicated flash-aware in-storage attention engine with KV cache management mechanisms to exploit the high internal bandwidths of CSDs instead of being limited by the PCIe bandwidth. The optimized P2P transmission between GPU and CSDs further reduces data migration overheads. Experimental results demonstrate that for a 13B model using an NVIDIA A6000 GPU, InstInfer improves throughput for long-sequence inference by up to 11.1times, compared to existing SSD-based solutions such as FlexGen.

Key-Value (KV) cache has become a bottleneck of LLMs for long-context generation. Despite the numerous efforts in this area, the optimization for the decoding phase is generally ignored. However, we believe such optimization is crucial, especially for long-output generation tasks based on the following two observations: (i) Excessive compression during the prefill phase, which requires specific full context impairs the comprehension of the reasoning task; (ii) Deviation of heavy hitters occurs in the reasoning tasks with long outputs. Therefore, SCOPE, a simple yet efficient framework that separately performs KV cache optimization during the prefill and decoding phases, is introduced. Specifically, the KV cache during the prefill phase is preserved to maintain the essential information, while a novel strategy based on sliding is proposed to select essential heavy hitters for the decoding phase. Memory usage and memory transfer are further optimized using adaptive and discontinuous strategies. Extensive experiments on LongGenBench show the effectiveness and generalization of SCOPE and its compatibility as a plug-in to other prefill-only KV compression methods.

Memory plays a key role in enhancing LLMs' performance when deployed to real-world applications. Existing solutions face trade-offs: explicit memory designs based on external storage require complex management and incur storage overhead, while implicit memory designs that store information via parameters struggle with reliable retrieval. In this paper, we propose R^3Mem, a memory network that optimizes both information Retention and Retrieval through Reversible context compression. Specifically, R^3Mem employs virtual memory tokens to compress and encode infinitely long histories, further enhanced by a hierarchical compression strategy that refines information from document- to entity-level for improved assimilation across granularities. For retrieval, R^3Mem employs a reversible architecture, reconstructing raw data by invoking the model backward with compressed information. Implemented via parameter-efficient fine-tuning, it can integrate seamlessly with any Transformer-based model. Experiments demonstrate that our memory design achieves state-of-the-art performance in long-context language modeling and retrieval-augmented generation tasks. It also significantly outperforms conventional memory modules in long-horizon interaction tasks like conversational agents, showcasing its potential for next-generation retrieval systems.

We present AutoCompress, a transformer compression method motivated by an empirical finding: in small transformers, Layer 0 carries disproportionately high task-critical information, with an NTK-based importance score of 3.6 compared to a maximum of 0.054 for all other layers -- a gap of over 60x. Based on this finding, we propose Critical Layer Isolation (CLI), an architecture that protects Layer 0 at full dimensionality, compresses all intermediate layers through a learned bottleneck, and restores the full dimension at the final layer. Applied to GPT-2 Medium (354.8M parameters), CLI-GPT2 achieves 204.5 perplexity on WikiText-103 with only 143.8M parameters -- a 2.47x compression ratio and 59.5% parameter reduction. Crucially, an ablation study demonstrates that a uniform bottleneck baseline of comparable size achieves only 571.8 perplexity under identical training conditions, confirming that the architectural decision to protect Layer 0 -- rather than simply reducing model size -- is the primary driver of performance. Code and checkpoints are publicly available.

Deploying large language models (LLMs) often faces challenges from substantial memory and computational costs. Quantization offers a solution, yet performance degradation in the sub-1-bit regime remains particularly difficult. This paper introduces LittleBit, a novel method for extreme LLM compression. It targets levels like 0.1 bits per weight (BPW), achieving nearly 31times memory reduction, e.g., Llama2-13B to under 0.9 GB. LittleBit represents weights in a low-rank form using latent matrix factorization, subsequently binarizing these factors. To counteract information loss from this extreme precision, it integrates a multi-scale compensation mechanism. This includes row, column, and an additional latent dimension that learns per-rank importance. Two key contributions enable effective training: Dual Sign-Value-Independent Decomposition (Dual-SVID) for stable quantization-aware training (QAT) initialization, and integrated Residual Compensation to mitigate errors. Extensive experiments confirm LittleBit's superiority in sub-1-bit quantization: e.g., its 0.1 BPW performance on Llama2-7B surpasses the leading method's 0.7 BPW. This establishes a superior size-performance trade-off, with kernel-level benchmarks indicating potential for a 5times speedup compared to FP16. LittleBit paves the way for deploying powerful LLMs in resource-constrained environments.

Visual Auto-Regressive (VAR) models significantly reduce inference steps through the "next-scale" prediction paradigm. However, progressive multi-scale generation incurs substantial memory overhead due to cumulative KV caching, limiting practical deployment. We observe a scale-depth asymmetric dependency in VAR: early scales exhibit extreme sensitivity to network depth, while later scales remain robust to depth reduction. Inspired by this, we propose VARiant: by equidistant sampling, we select multiple subnets ranging from 16 to 2 layers from the original 30-layer VAR-d30 network. Early scales are processed by the full network, while later scales utilize subnet. Subnet and the full network share weights, enabling flexible depth adjustment within a single model. However, weight sharing between subnet and the entire network can lead to optimization conflicts. To address this, we propose a progressive training strategy that breaks through the Pareto frontier of generation quality for both subnets and the full network under fixed-ratio training, achieving joint optimality. Experiments on ImageNet demonstrate that, compared to the pretrained VAR-d30 (FID 1.95), VARiant-d16 and VARiant-d8 achieve nearly equivalent quality (FID 2.05/2.12) while reducing memory consumption by 40-65%. VARiant-d2 achieves 3.5 times speedup and 80% memory reduction at moderate quality cost (FID 2.97). In terms of deployment, VARiant's single-model architecture supports zero-cost runtime depth switching and provides flexible deployment options from high quality to extreme efficiency, catering to diverse application scenarios.

Large language model fine-tuning is bottlenecked by memory: a 7B parameter model requires 84GB--14GB for weights, 14GB for gradients, and 56GB for FP32 optimizer states--exceeding even A100-40GB capacity. We present Chronicals, an open-source training framework achieving 3.51x speedup over Unsloth through four synergistic optimizations: (1) fused Triton kernels eliminating 75% of memory traffic via RMSNorm (7x), SwiGLU (5x), and QK-RoPE (2.3x) fusion; (2) Cut Cross-Entropy reducing logit memory from 5GB to 135MB through online softmax computation; (3) LoRA+ with theoretically-derived 16x differential learning rates between adapter matrices; and (4) Best-Fit Decreasing sequence packing recovering 60-75% of compute wasted on padding. On Qwen2.5-0.5B with A100-40GB, Chronicals achieves 41,184 tokens/second for full fine-tuning versus Unsloth's 11,736 tokens/second (3.51x). For LoRA at rank 32, we reach 11,699 tokens/second versus Unsloth MAX's 2,857 tokens/second (4.10x). Critically, we discovered that Unsloth's reported 46,000 tokens/second benchmark exhibited zero gradient norms--the model was not training. We provide complete mathematical foundations: online softmax correctness proofs, FlashAttention IO complexity bounds O(N^2 d^2 M^{-1}), LoRA+ learning rate derivations from gradient magnitude analysis, and bin-packing approximation guarantees. All implementations, benchmarks, and proofs are available at https://github.com/Ajwebdevs/Chronicals with pip installation via https://pypi.org/project/chronicals/.

Serving numerous users and requests concurrently requires good fairness in Large Language Models (LLMs) serving system. This ensures that, at the same cost, the system can meet the Service Level Objectives (SLOs) of more users , such as time to first token (TTFT) and time between tokens (TBT), rather than allowing a few users to experience performance far exceeding the SLOs. To achieve better fairness, the preemption-based scheduling policy dynamically adjusts the priority of each request to maintain balance during runtime. However, existing systems tend to overly prioritize throughput, overlooking the overhead caused by preemption-induced context switching, which is crucial for maintaining fairness through priority adjustments. In this work, we identify three main challenges that result in this overhead. 1) Inadequate I/O utilization. 2) GPU idleness. 3) Unnecessary I/O transmission during multi-turn conversations. Our key insight is that the block-based KV cache memory policy in existing systems, while achieving near-zero memory waste, leads to discontinuity and insufficient granularity in the KV cache memory. To respond, we introduce FastSwitch, a fairness-aware serving system that not only aligns with existing KV cache memory allocation policy but also mitigates context switching overhead. Our evaluation shows that FastSwitch outperforms the state-of-the-art LLM serving system vLLM with speedups of 1.4-11.2x across different tail TTFT and TBT.

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.

We prove that any Turing machine running on inputs of arbitrary length can be simulated by a constant bit-size transformer, as long as the context window is sufficiently long. This improves previous works, which require scaling up either the model's precision or the number of parameters on longer inputs. Furthermore, we prove that the complexity class SPACE[s(n)] exactly characterizes the expressive power of a constant bit-size transformer with a context window of length s(n). Our approach relies on simulating Post machines, a Turing-complete computational model. Post machines can be modeled as automata equipped with a queue, exhibiting computational behaviors naturally aligned with those of transformers. The behavioral similarity between transformers and Post machines may offer new insights into the mechanisms underlying the reasoning abilities of transformers.

Large reasoning models (LRMs) excel on complex problems but face a critical barrier to efficiency: reinforcement learning (RL) training requires long rollouts for outcome-based rewards, where autoregressive decoding dominates time and memory usage. While sliding-window cache strategies can bound memory, they disrupt long-context reasoning and degrade performance. We introduce Progressive Thought Encoding, a parameter-efficient fine-tuning method that enables LRMs to reason effectively under fixed-size caches. By progressively encoding intermediate reasoning into fixed-size vector representations, our approach eliminates the need to backpropagate through full-cache rollouts, thereby reducing memory usage, while maintaining constant memory during inference. Experiments on three models, including Qwen2.5-3B-Instruct, Qwen2.5-7B-Instruct, and DeepSeek-R1-Distill-Llama-8B, on six widely used challenging mathematical benchmarks show consistent gains: our method achieves +19.3% improvement over LoRA-based fine-tuning and +29.9% over LRMs without fine-tuning on average, with up to +23.4 accuracy improvement on AIME2024/2025 under the same tight cache budgets. These results demonstrate that Progressive Thought Encoding not only improves reasoning accuracy but also makes RL training of LRMs substantially more efficient and scalable under real-world memory constraints.

Large Language Model (LLM) inference in production must meet stringent service-level objectives for both time-to-first-token (TTFT) and time-between-token (TBT) while maximizing throughput under fixed compute, memory, and interconnect budgets. Modern serving systems adopt stall-free scheduling techniques such as chunked prefill, which splits long prompt processing along the token dimension and interleaves prefill with ongoing decode iterations. While effective at stabilizing TBT, chunked prefill incurs substantial overhead in Mixture-of-Experts (MoE) models: redundant expert weight loads increase memory traffic by up to 39% and inflate energy consumption. We propose layered prefill, a new scheduling paradigm that treats transformer layer groups as the primary scheduling unit. By vertically partitioning the model into contiguous layer groups and interleaving prefill and decode across the groups, layered prefill sustains stall-free decoding while eliminating chunk-induced MoE weight reloads. It reduces off-chip bandwidth demand, lowering TTFT by up to 70%, End-to-End latency by 41% and per-token energy by up to 22%. Evaluations show that layered prefill consistently improves the TTFT--TBT Pareto frontier over chunked prefill, reducing expert-load traffic and energy cost while maintaining stall-free decoding. Overall, shifting the scheduling axis from tokens to layers unlocks a new operating regime for high-efficiency, energy-aware LLM serving in co-located environments.

Mixture-of-Experts (MoE) enables efficient scaling of large language models (LLMs) with sparsely activated experts during inference. To effectively deploy large MoE models on memory-constrained devices, many systems introduce *expert offloading* that caches a subset of experts in fast memory, leaving others on slow memory to run on CPU or load on demand. While some research has exploited the locality of expert activations, where consecutive tokens activate similar experts, the degree of this **local routing consistency** varies across models and remains understudied. In this paper, we propose two metrics to measure local routing consistency of MoE models: (1) **Segment Routing Best Performance (SRP)**, which evaluates how well a fixed group of experts can cover the needs of a segment of tokens, and (2) **Segment Cache Best Hit Rate (SCH)**, which measures the optimal segment-level cache hit rate under a given cache size limit. We analyzed 20 MoE LLMs with diverse sizes and architectures and found that models that apply MoE on every layer and do not use shared experts exhibit the highest local routing consistency. We further showed that domain-specialized experts contribute more to routing consistency than vocabulary-specialized ones, and that most models can balance between cache effectiveness and efficiency with cache sizes approximately 2x the active experts. These findings pave the way for memory-efficient MoE design and deployment without compromising inference speed. We publish the code for replicating experiments at https://github.com/ljcleo/moe-lrc .

Training large-scale models under given resources requires careful design of parallelism strategies. In particular, the efficiency notion of critical batch size (CBS), concerning the compromise between time and compute, marks the threshold beyond which greater data parallelism leads to diminishing returns. To operationalize it, we propose a measure of CBS and pre-train a series of auto-regressive language models, ranging from 85 million to 1.2 billion parameters, on the C4 dataset. Through extensive hyper-parameter sweeps and careful control of factors such as batch size, momentum, and learning rate along with its scheduling, we systematically investigate the impact of scale on CBS. Then we fit scaling laws with respect to model and data sizes to decouple their effects. Overall, our results demonstrate that CBS scales primarily with data size rather than model size, a finding we justify theoretically through the analysis of infinite-width limits of neural networks and infinite-dimensional least squares regression. Of independent interest, we highlight the importance of common hyper-parameter choices and strategies for studying large-scale pre-training beyond fixed training durations.

Large Language Models (LLMs) have become an essential infrastructure for Artificial General Intelligence (AGI), yet their lack of well-defined memory management systems hinders the development of long-context reasoning, continual personalization, and knowledge consistency.Existing models mainly rely on static parameters and short-lived contextual states, limiting their ability to track user preferences or update knowledge over extended periods.While Retrieval-Augmented Generation (RAG) introduces external knowledge in plain text, it remains a stateless workaround without lifecycle control or integration with persistent representations.Recent work has modeled the training and inference cost of LLMs from a memory hierarchy perspective, showing that introducing an explicit memory layer between parameter memory and external retrieval can substantially reduce these costs by externalizing specific knowledge. Beyond computational efficiency, LLMs face broader challenges arising from how information is distributed over time and context, requiring systems capable of managing heterogeneous knowledge spanning different temporal scales and sources. To address this challenge, we propose MemOS, a memory operating system that treats memory as a manageable system resource. It unifies the representation, scheduling, and evolution of plaintext, activation-based, and parameter-level memories, enabling cost-efficient storage and retrieval. As the basic unit, a MemCube encapsulates both memory content and metadata such as provenance and versioning. MemCubes can be composed, migrated, and fused over time, enabling flexible transitions between memory types and bridging retrieval with parameter-based learning. MemOS establishes a memory-centric system framework that brings controllability, plasticity, and evolvability to LLMs, laying the foundation for continual learning and personalized modeling.

The demand for inference on extremely large scale LLMs has seen enormous growth in the recent months. It made evident the colossal shortage of dedicated hardware capable of efficient and fast processing of the involved compute and memory movement. The problem is aggravated by the exploding raise in the lengths of the sequences being processed, since those require efficient on-chip storage of the KV-cache of size proportional to the sequence length. To make the required compute feasible and fit the involved data into available memory, numerous quantization techniques have been proposed that allow accurate quantization for both weights and activations. One of the main recent breakthroughs in this direction was introduction of the family of Block Floating Point (BFP) formats characterized by a block of mantissas with a shared scale factor. These enable memory- power-, and compute- efficient hardware support of the tensor operations and provide extremely good quantization accuracy. The main issues preventing widespread application of block formats is caused by the presence of outliers in weights and activations since those affect the accuracy of the other values in the same block. In this paper, we focus on the most critical problem of limited KV-cache storage. We propose a novel approach enabling usage of low precision BFP formats without compromising the resulting model accuracy. We exploit the common channel-wise patterns exhibited by the outliers to rearrange them in such a way, that their quantization quality is significantly improved. The methodology yields 2x savings in the memory footprint without significant degradation of the model's accuracy. Importantly, the rearrangement of channels happens at the compile time and thus has no impact on the inference latency.

Training billion-parameter models requires distributing model states across GPUs using fully sharded data parallel (i.e., ZeRO-3). While ZeRO-3 succeeds on clusters with high-bandwidth NVLink and InfiniBand interconnects, researchers with commodity hardware face severe inter-node all-gather bottlenecks. Existing optimizations take two approaches: GPU memory caching (MiCS, ZeRO++) trades memory capacity for reduced communication, triggering out-of-memory failures on large models; host memory offloading (ZeRO-Offload, ZeRO-Infinity) extends capacity but degrades throughput due to PCIe overhead. We observe that on bandwidth-limited clusters, host memory can serve not as an overflow tier but as a fast caching layer that outperforms inter-node communication. Based on this insight, we propose FCDP, which eliminates redundant inter-node communication while preserving ZeRO-3's minimal GPU memory footprint. FCDP caches forward-pass parameters in host memory and reuses them during the backward pass via fast intra-node all-gather, reducing inter-node all-gather by 50%. For parameter-efficient fine-tuning (PEFT), FCDP selectively communicates only trainable parameters to maximize caching, reducing inter-node traffic by over 99%. In our commodity cluster setup, FCDP achieves up to 100x higher throughput than ZeRO-3 and 51x higher than ZeRO++, while maintaining ZeRO-3's maximum batch size.

Large Language Models (LLMs) have grown rapidly in size, creating significant challenges for efficient deployment on resource-constrained hardware. In this paper, we introduce Dynamic-Length Float (DFloat11), a lossless compression framework that reduces LLM size by 30% while preserving outputs that are bit-for-bit identical to the original model. DFloat11 is motivated by the low entropy in the BFloat16 weight representation of LLMs, which reveals significant inefficiency in existing storage format. By applying entropy coding, DFloat11 assigns dynamic-length encodings to weights based on frequency, achieving near information-optimal compression without any loss of precision. To facilitate efficient inference with dynamic-length encodings, we develop a custom GPU kernel for fast online decompression. Our design incorporates the following: (i) decomposition of memory-intensive lookup tables (LUTs) into compact LUTs that fit in GPU SRAM, (ii) a two-phase kernel for coordinating thread read/write positions using lightweight auxiliary variables, and (iii) transformer-block-level decompression to minimize latency. Experiments on recent models, including Llama-3.1, Qwen-2.5, and Gemma-3, validates our hypothesis that DFloat11 achieves around 30% model size reduction while preserving bit-for-bit exact outputs. Compared to a potential alternative of offloading parts of an uncompressed model to the CPU to meet memory constraints, DFloat11 achieves 1.9-38.8x higher throughput in token generation. With a fixed GPU memory budget, DFloat11 enables 5.3-13.17x longer context lengths than uncompressed models. Notably, our method enables lossless inference of Llama-3.1-405B, an 810GB model, on a single node equipped with 8x80GB GPUs. Our code and models are available at https://github.com/LeanModels/DFloat11.

Large Language Models (LLMs) have demonstrated strong capabilities across various domains, with recent advancements in challenging reasoning tasks such as mathematics and programming. However, solving reasoning tasks often requires long decoding chains (of thoughts), which incur O(N) time and memory consumption, where N is the chain length. To mitigate O(N) time and memory consumption, existing sparsity-based algorithms propose retaining only the most critical token's intermediate data (i.e., key-value cache) and discarding the rest. However, these existing algorithms struggle with the ``impossible trinity'' of accuracy, time, and memory. For example, the state-of-the-art algorithm, Quest, achieves high accuracy with O(L) time but O(N) memory (L is the cache budget, L ll N). To address this issue, in this paper, we identify a new attention pattern during the decode stage of reasoning tasks, where milestone tokens (analogous to lemmas in mathematical proofs) emerge, are utilized, and then become unimportant afterward. Based on this pattern, we propose a new algorithm named RaaS that identifies and retains milestone tokens only until they are no longer needed, achieving high accuracy with O(L) time and O(L) memory complexity.

The availability of unprecedented unsupervised training data, along with neural scaling laws, has resulted in an unprecedented surge in model size and compute requirements for serving/training LLMs. However, the main performance bottleneck is increasingly shifting to memory bandwidth. Over the past 20 years, peak server hardware FLOPS has been scaling at 3.0x/2yrs, outpacing the growth of DRAM and interconnect bandwidth, which have only scaled at 1.6 and 1.4 times every 2 years, respectively. This disparity has made memory, rather than compute, the primary bottleneck in AI applications, particularly in serving. Here, we analyze encoder and decoder Transformer models and show how memory bandwidth can become the dominant bottleneck for decoder models. We argue for a redesign in model architecture, training, and deployment strategies to overcome this memory limitation.

To improve the efficiency of distributed large language model (LLM) inference, various parallelization strategies, such as tensor and pipeline parallelism, have been proposed. However, the distinct computational characteristics inherent in the two stages of LLM inference-prefilling and decoding-render a single static parallelization strategy insufficient for the effective optimization of both stages. In this work, we present Seesaw, an LLM inference engine optimized for throughput-oriented tasks. The key idea behind Seesaw is dynamic model re-sharding, a technique that facilitates the dynamic reconfiguration of parallelization strategies across stages, thereby maximizing throughput at both phases. To mitigate re-sharding overhead and optimize computational efficiency, we employ tiered KV cache buffering and transition-minimizing scheduling. These approaches work synergistically to reduce the overhead caused by frequent stage transitions while ensuring maximum batching efficiency. Our evaluation demonstrates that Seesaw achieves a throughput increase of up to 1.78x (1.36x on average) compared to vLLM, the most widely used state-of-the-art LLM inference engine.

Standard mixed-precision training of neural networks requires many bytes of accelerator memory for each model parameter. These bytes reflect not just the parameter itself, but also its gradient and one or more optimizer state variables. With each of these values typically requiring 4 bytes, training even a 7 billion parameter model can be impractical for researchers with less than 100GB of accelerator memory. We introduce FlashOptim, a suite of optimizations that reduces per-parameter memory by over 50% while preserving model quality and API compatibility. Our approach introduces two key techniques. First, we improve master weight splitting by finding and exploiting a tight bound on its quantization error. Second, we design companding functions that greatly reduce the error in 8-bit optimizer state quantization. Together with 16-bit gradients, these techniques reduce AdamW memory from 16 bytes to 7 bytes per parameter, or 5 bytes with gradient release. They also cut model checkpoint sizes by more than half. Experiments with FlashOptim applied to SGD, AdamW, and Lion show no measurable quality degradation on any task from a collection of standard vision and language benchmarks, including Llama-3.1-8B finetuning.

Mixture-of-Experts (MoE) architectures in large language models (LLMs) deliver exceptional performance and reduced inference costs compared to dense LLMs. However, their large parameter counts result in prohibitive memory requirements, limiting practical deployment. While existing pruning methods primarily focus on expert-level pruning, this coarse granularity often leads to substantial accuracy degradation. In this work, we introduce HEAPr, a novel pruning algorithm that decomposes experts into smaller, indivisible atomic experts, enabling more precise and flexible atomic expert pruning. To measure the importance of each atomic expert, we leverage second-order information based on principles similar to Optimal Brain Surgeon (OBS) theory. To address the computational and storage challenges posed by second-order information, HEAPr exploits the inherent properties of atomic experts to transform the second-order information from expert parameters into that of atomic expert parameters, and further simplifies it to the second-order information of atomic expert outputs. This approach reduces the space complexity from O(d^4), where d is the model's dimensionality, to O(d^2). HEAPr requires only two forward passes and one backward pass on a small calibration set to compute the importance of atomic experts. Extensive experiments on MoE models, including DeepSeek MoE and Qwen MoE family, demonstrate that HEAPr outperforms existing expert-level pruning methods across a wide range of compression ratios and benchmarks. Specifically, HEAPr achieves nearly lossless compression at compression ratios of 20% ~ 25% in most models, while also reducing FLOPs nearly by 20%. The code can be found at https://github.com/LLIKKE/HEAPr{https://github.com/LLIKKE/HEAPr}.

Linear sequence modeling methods, such as linear attention, state space modeling, and linear RNNs, offer significant efficiency improvements by reducing the complexity of training and inference. However, these methods typically compress the entire input sequence into a single fixed-size memory state, which leads to suboptimal performance on recall-intensive downstream tasks. Drawing inspiration from neuroscience, particularly the brain's ability to maintain robust long-term memory while mitigating "memory interference", we introduce a novel architecture called Mixture-of-Memories (MoM). MoM utilizes multiple independent memory states, with a router network directing input tokens to specific memory states. This approach greatly enhances the overall memory capacity while minimizing memory interference. As a result, MoM performs exceptionally well on recall-intensive tasks, surpassing existing linear sequence modeling techniques. Despite incorporating multiple memory states, the computation of each memory state remains linear in complexity, allowing MoM to retain the linear-complexity advantage during training, while constant-complexity during inference. Our experimental results show that MoM significantly outperforms current linear sequence models on downstream language tasks, particularly recall-intensive tasks, and even achieves performance comparable to Transformer models. The code is released at https://github.com/OpenSparseLLMs/MoM and is also released as a part of https://github.com/OpenSparseLLMs/Linear-MoE.

Mixture-of-Experts (MoE) architectures offer a general solution to the high inference costs of large language models (LLMs) via sparse routing, bringing faster and more accurate models, at the cost of massive parameter counts. For example, the SwitchTransformer-c2048 model has 1.6 trillion parameters, requiring 3.2TB of accelerator memory to run efficiently, which makes practical deployment challenging and expensive. In this paper, we present a solution to this memory problem, in form of a new compression and execution framework called QMoE. Specifically, QMoE consists of a scalable algorithm which accurately compresses trillion-parameter MoEs to less than 1 bit per parameter, in a custom format co-designed with bespoke GPU decoding kernels to facilitate efficient end-to-end compressed inference, with minor runtime overheads relative to uncompressed execution. Concretely, QMoE can compress the 1.6 trillion parameter SwitchTransformer-c2048 model to less than 160GB (20x compression, 0.8 bits per parameter) at only minor accuracy loss, in less than a day on a single GPU. This enables, for the first time, the execution of a trillion-parameter model on affordable commodity hardware, like a single server with 4x NVIDIA A6000 or 8x NVIDIA 3090 GPUs, at less than 5% runtime overhead relative to ideal uncompressed inference. The source code and compressed models are available at github.com/IST-DASLab/qmoe.

Large Language Models (LLMs) face a crucial challenge from fixed context windows and inadequate memory management, leading to a severe shortage of long-term memory capabilities and limited personalization in the interactive experience with AI agents. To overcome this challenge, we innovatively propose a Memory Operating System, i.e., MemoryOS, to achieve comprehensive and efficient memory management for AI agents. Inspired by the memory management principles in operating systems, MemoryOS designs a hierarchical storage architecture and consists of four key modules: Memory Storage, Updating, Retrieval, and Generation. Specifically, the architecture comprises three levels of storage units: short-term memory, mid-term memory, and long-term personal memory. Key operations within MemoryOS include dynamic updates between storage units: short-term to mid-term updates follow a dialogue-chain-based FIFO principle, while mid-term to long-term updates use a segmented page organization strategy. Our pioneering MemoryOS enables hierarchical memory integration and dynamic updating. Extensive experiments on the LoCoMo benchmark show an average improvement of 49.11% on F1 and 46.18% on BLEU-1 over the baselines on GPT-4o-mini, showing contextual coherence and personalized memory retention in long conversations. The implementation code is open-sourced at https://github.com/BAI-LAB/MemoryOS.

Current multi-agent Large Language Model (LLM) frameworks suffer from linear memory scaling, rendering "System 2" parallel reasoning impractical on consumer hardware. We present Warp Cortex, an asynchronous architecture that theoretically enables million-agent cognitive scaling by decoupling agent logic from physical memory. Through Singleton Weight Sharing and a novel Topological Synapse--inspired by hybrid landmarking techniques from Topological Data Analysis (TDA)--we reduce memory complexity from O(N * L) to O(1) for weights and O(N * k) for context, where k << L. By treating the KV-cache as a point cloud in latent space, we apply witness-complex-inspired sparsification to preserve persistent homological features of the context manifold. On a single NVIDIA RTX 4090, we empirically demonstrate 100 concurrent agents at 2.2 GB total VRAM, with theoretical capacity exceeding 1,000 agents before compute latency becomes the bottleneck. We further introduce Referential Injection, a non-intrusive KV-cache update mechanism that allows asynchronous sub-agents to influence primary generation without stream disruption.

It is widely acknowledged that the performance of Transformer models is exponentially related to their number of parameters and computational complexity. While approaches like Mixture of Experts (MoE) decouple parameter count from computational complexity, they still face challenges in inference due to high memory access costs. This work introduces UltraMem, incorporating large-scale, ultra-sparse memory layer to address these limitations. Our approach significantly reduces inference latency while maintaining model performance. We also investigate the scaling laws of this new architecture, demonstrating that it not only exhibits favorable scaling properties but outperforms traditional models. In our experiments, we train networks with up to 20 million memory slots. The results show that our method achieves state-of-the-art inference speed and model performance within a given computational budget.

This paper presents a simple yet effective approach that improves continual test-time adaptation (TTA) in a memory-efficient manner. TTA may primarily be conducted on edge devices with limited memory, so reducing memory is crucial but has been overlooked in previous TTA studies. In addition, long-term adaptation often leads to catastrophic forgetting and error accumulation, which hinders applying TTA in real-world deployments. Our approach consists of two components to address these issues. First, we present lightweight meta networks that can adapt the frozen original networks to the target domain. This novel architecture minimizes memory consumption by decreasing the size of intermediate activations required for backpropagation. Second, our novel self-distilled regularization controls the output of the meta networks not to deviate significantly from the output of the frozen original networks, thereby preserving well-trained knowledge from the source domain. Without additional memory, this regularization prevents error accumulation and catastrophic forgetting, resulting in stable performance even in long-term test-time adaptation. We demonstrate that our simple yet effective strategy outperforms other state-of-the-art methods on various benchmarks for image classification and semantic segmentation tasks. Notably, our proposed method with ResNet-50 and WideResNet-40 takes 86% and 80% less memory than the recent state-of-the-art method, CoTTA.

High throughput serving of large language models (LLMs) requires batching sufficiently many requests at a time. However, existing systems struggle because the key-value cache (KV cache) memory for each request is huge and grows and shrinks dynamically. When managed inefficiently, this memory can be significantly wasted by fragmentation and redundant duplication, limiting the batch size. To address this problem, we propose PagedAttention, an attention algorithm inspired by the classical virtual memory and paging techniques in operating systems. On top of it, we build vLLM, an LLM serving system that achieves (1) near-zero waste in KV cache memory and (2) flexible sharing of KV cache within and across requests to further reduce memory usage. Our evaluations show that vLLM improves the throughput of popular LLMs by 2-4times with the same level of latency compared to the state-of-the-art systems, such as FasterTransformer and Orca. The improvement is more pronounced with longer sequences, larger models, and more complex decoding algorithms. vLLM's source code is publicly available at https://github.com/vllm-project/vllm

AI memory systems are evolving toward unified context layers that enable efficient cross-agent collaboration and multi-tool workflows, facilitating better accumulation of personal data and learning of user preferences. However, centralization creates a trust crisis where users must entrust cloud providers with sensitive digital memory data. We identify a core tension between personalization demands and data sovereignty: centralized memory systems enable efficient cross-agent collaboration but expose users' sensitive data to cloud provider risks, while private deployments provide security but limit collaboration. To resolve this tension, we aim to achieve local-equivalent security while enabling superior maintenance efficiency and collaborative capabilities. We propose a five-layer architecture abstracting common functional components of AI memory systems: Storage, Extraction, Learning, Retrieval, and Governance. By applying TEE protection to each layer, we establish a trustworthy framework. Based on this, we design MemTrust, a hardware-backed zero-trust architecture that provides cryptographic guarantees across all layers. Our contributions include the five-layer abstraction, "Context from MemTrust" protocol for cross-application sharing, side-channel hardened retrieval with obfuscated access patterns, and comprehensive security analysis. The architecture enables third-party developers to port existing systems with acceptable development costs, achieving system-wide trustworthiness. We believe that AI memory plays a crucial role in enhancing the efficiency and collaboration of agents and AI tools. AI memory will become the foundational infrastructure for AI agents, and MemTrust serves as a universal trusted framework for AI memory systems, with the goal of becoming the infrastructure of memory infrastructure.

Many computational factors limit broader deployment of large language models. In this paper, we focus on a memory bottleneck imposed by the key-value (KV) cache, a computational shortcut that requires storing previous KV pairs during decoding. While existing KV cache methods approach this problem by pruning or evicting large swaths of relatively less important KV pairs to dramatically reduce the memory footprint of the cache, they can have limited success in tasks that require recollecting a majority of previous tokens. To alleviate this issue, we propose LESS, a simple integration of a (nearly free) constant sized cache with eviction-based cache methods, such that all tokens can be queried at later decoding steps. Its ability to retain information throughout time shows merit on a variety of tasks where we demonstrate LESS can help reduce the performance gap from caching everything, sometimes even matching it, all while being efficient.

Large Language Models (LLMs) are fundamentally constrained by the quadratic computational cost of self-attention and the "Lost in the Middle" phenomenon, where reasoning capabilities degrade as context windows expand. Existing solutions, primarily "Flat RAG" architectures relying on vector databases, treat memory as an unstructured bag of embeddings, failing to capture the hierarchical and temporal structure of long-horizon interactions. This paper presents Aeon, a Neuro-Symbolic Cognitive Operating System that redefines memory as a managed OS resource. Aeon structures memory into a Memory Palace (a spatial index implemented via Atlas, a SIMD-accelerated Page-Clustered Vector Index) and a Trace (a neuro-symbolic episodic graph). This architecture introduces three advances: (1) Symmetric INT8 Scalar Quantization, achieving 3.1x spatial compression and 5.6x math acceleration via NEON SDOT intrinsics; (2) a decoupled Write-Ahead Log (WAL) ensuring crash-recoverability with statistically negligible overhead (<1%); and (3) a Sidecar Blob Arena eliminating the prior 440-character text ceiling via an append-only mmap-backed blob file with generational garbage collection. The Semantic Lookaside Buffer (SLB) exploits conversational locality to achieve sub-5us retrieval latencies, with INT8 vectors dequantized to FP32 on cache insertion to preserve L1-resident lookup performance. Benchmarks on Apple M4 Max demonstrate that the combined architecture achieves 4.70ns INT8 dot product latency, 3.09us tree traversal at 100K nodes (3.4x over FP32), and P99 read latency of 750ns under hostile 16-thread contention via epoch-based reclamation.

Despite a century of empirical memory research, existing AI agent memory systems rely on system-engineering metaphors (virtual-memory paging, flat LLM storage, Zettelkasten notes), none integrating principles of consolidation, forgetting, and reconsolidation. We present ZenBrain, a multi-layer memory architecture integrating fifteen neuroscience models. It implements seven memory layers (working, short-term, episodic, semantic, procedural, core, cross-context) orchestrated by nine foundational algorithms (Two-Factor Synaptic Model, vmPFC-coupled FSRS, Simulation-Selection sleep, Bayesian confidence, and five more) plus six new Predictive Memory Architecture (PMA) components: a four-channel NeuromodulatorEngine, prediction-error-gated ReconsolidationEngine, TripleCopyMemory with divergent decay, four-dimensional PriorityMap with amygdala fast-path, StabilityProtector (NogoA/HDAC3 analogue), and MetacognitiveMonitor for bias detection. The 15-algorithm ablation reveals a cooperative survival network: under stress, 9 of 15 algorithms become individually critical (delta-Q up to -93.7%, Wilcoxon, 10 seeds, alpha=0.005). Simulation-Selection sleep achieves 37% stability improvement (p<0.005) with 47.4% storage reduction. TripleCopyMemory retains S(t)=0.912 at 30 days; PriorityMap reaches NDCG@10=0.997. Multi-layer routing beats a flat single-layer baseline by 20.7% F1 on LoCoMo (p<0.005) and 19.5% on MemoryArena (p=0.015). On LongMemEval-500, ZenBrain holds the highest mean rank on all 12 system-judge cells (4 systems x 3 LLM judges), three-judge mean J=0.545 vs letta=0.485, a-mem=0.414, mem0=0.394; all 9 pair-wise contrasts clear Bonferroni (alpha=0.05/18, min p=6.2e-31, d in [0.18, 0.52]). Under LongMemEval's binary judge, ZenBrain reaches 91.3% of oracle accuracy at 1/106th the per-query token budget. Open-source with 11,589 automated test cases.

The growing prevalence of Large Language Models (LLMs) and their substantial memory requirements have prompted renewed interest in CPU offloading as a method to compensate for limited GPU memory. In particular, when CPU memory is leveraged to temporarily store intermediate states of LLMs, CPU memory becomes a new bottleneck and soon reaches the capacity limitation of commodity CPUs. In this work, we investigate the effectiveness of Compute Express Link (CXL) add-in card (AIC) memory as an extension to CPU memory, enabling larger model sizes and longer context lengths during fine-tuning. Through extensive benchmarking, this study quantifies the performance overhead introduced by transferring data between CXL memory, CPU, and GPUs, focusing on how concurrency and data volume influence bandwidth utilization and latency. This study also compares CPUbased optimizer steps when model parameters, gradients, and optimizer states reside in local memory versus CXL memory, revealing that naive adoption of CXL often degrades performance during the optimizer phase. To overcome these challenges, this study proposes a CXL-aware allocation to strategically partition CPU offloading workloads across both local and CXL memory. This study further demonstrates that employing multiple AICs significantly reduces bandwidth contention, thus improving scalability. Experimental results show that these optimizations enable efficient long-context LLM fine-tuning, underscoring CXL as a promising avenue for unlocking the full potential of CPU offloading in long-context LLM fine-tuning.

Agentic memory systems have become critical for enabling LLM agents to maintain long-term context and retrieve relevant information efficiently. However, existing memory frameworks suffer from a fundamental limitation: they perform exhaustive retrieval across the entire storage layer regardless of query characteristics. This brute-force approach creates severe latency bottlenecks as memory grows, hindering real-time agent interactions. We propose SwiftMem, a query-aware agentic memory system that achieves sub-linear retrieval through specialized indexing over temporal and semantic dimensions. Our temporal index enables logarithmic-time range queries for time-sensitive retrieval, while the semantic DAG-Tag index maps queries to relevant topics through hierarchical tag structures. To address memory fragmentation during growth, we introduce an embedding-tag co-consolidation mechanism that reorganizes storage based on semantic clusters to improve cache locality. Experiments on LoCoMo and LongMemEval benchmarks demonstrate that SwiftMem achieves 47times faster search compared to state-of-the-art baselines while maintaining competitive accuracy, enabling practical deployment of memory-augmented LLM agents.

Recent works have highlighted the significance of memory mechanisms in LLM-based agents, which enable them to store observed information and adapt to dynamic environments. However, evaluating their memory capabilities still remains challenges. Previous evaluations are commonly limited by the diversity of memory levels and interactive scenarios. They also lack comprehensive metrics to reflect the memory capabilities from multiple aspects. To address these problems, in this paper, we construct a more comprehensive dataset and benchmark to evaluate the memory capability of LLM-based agents. Our dataset incorporates factual memory and reflective memory as different levels, and proposes participation and observation as various interactive scenarios. Based on our dataset, we present a benchmark, named MemBench, to evaluate the memory capability of LLM-based agents from multiple aspects, including their effectiveness, efficiency, and capacity. To benefit the research community, we release our dataset and project at https://github.com/import-myself/Membench.

Large language models (LLMs) have emerged due to their capability to generate high-quality content across diverse contexts. To reduce their explosively increasing demands for computing resources, a mixture of experts (MoE) has emerged. The MoE layer enables exploiting a huge number of parameters with less computation. Applying state-of-the-art continuous batching increases throughput; however, it leads to frequent DRAM access in the MoE and attention layers. We observe that conventional computing devices have limitations when processing the MoE and attention layers, which dominate the total execution time and exhibit low arithmetic intensity (Op/B). Processing MoE layers only with devices targeting low-Op/B such as processing-in-memory (PIM) architectures is challenging due to the fluctuating Op/B in the MoE layer caused by continuous batching. To address these challenges, we propose Duplex, which comprises xPU tailored for high-Op/B and Logic-PIM to effectively perform low-Op/B operation within a single device. Duplex selects the most suitable processor based on the Op/B of each layer within LLMs. As the Op/B of the MoE layer is at least 1 and that of the attention layer has a value of 4-8 for grouped query attention, prior PIM architectures are not efficient, which place processing units inside DRAM dies and only target extremely low-Op/B (under one) operations. Based on recent trends, Logic-PIM adds more through-silicon vias (TSVs) to enable high-bandwidth communication between the DRAM die and the logic die and place powerful processing units on the logic die, which is best suited for handling low-Op/B operations ranging from few to a few dozens. To maximally utilize the xPU and Logic-PIM, we propose expert and attention co-processing.

The key-value (KV) cache is a foundational optimization in Transformer-based large language models (LLMs), eliminating redundant recomputation of past token representations during autoregressive generation. However, its memory footprint scales linearly with context length, imposing critical bottlenecks on GPU memory capacity, memory bandwidth, and inference throughput as production LLMs push context windows from thousands to millions of tokens. Efficient KV cache management has thus become a first-order challenge for scalable LLM deployment. This paper provides a systematic review of recent KV cache optimization techniques, organizing them into five principal directions: cache eviction, cache compression, hybrid memory solutions, novel attention mechanisms, and combination strategies. For each category we analyze the underlying mechanisms, deployment trade-offs, and empirical performance across memory reduction, throughput, and model accuracy metrics. We further map techniques to seven practical deployment scenarios, including long-context single requests, high-throughput datacenter serving, edge devices, multi-turn conversations, and accuracy-critical reasoning, providing actionable guidance for practitioners selecting among competing approaches. Our analysis reveals that no single technique dominates across all settings; instead, the optimal strategy depends on context length, hardware constraints, and workload characteristics, pointing toward adaptive, multi-stage optimization pipelines as a promising direction for future research.

Mixture of Experts (MoE) architectures have significantly increased computational efficiency in both research and real-world applications of large-scale machine learning models. However, their scalability and efficiency under memory constraints remain relatively underexplored. In this work, we present joint scaling laws for dense and MoE models, incorporating key factors such as the number of active parameters, dataset size, and the number of experts. Our findings provide a principled framework for selecting the optimal MoE configuration under fixed memory and compute budgets. Surprisingly, we show that MoE models can be more memory-efficient than dense models, contradicting conventional wisdom. To derive and validate the theoretical predictions of our scaling laws, we conduct over 280 experiments with up to 2.7B active parameters and up to 5B total parameters. These results offer actionable insights for designing and deploying MoE models in practical large-scale training scenarios.

We introduce GoldFinch, a hybrid Linear Attention/Transformer sequence model that uses a new technique to efficiently generate a highly compressed and reusable KV-Cache in linear time and space with respect to sequence length. GoldFinch stacks our new GOLD transformer on top of an enhanced version of the Finch (RWKV-6) architecture. We train up to 1.5B parameter class models of the Finch, Llama, and GoldFinch architectures, and find dramatically improved modeling performance relative to both Finch and Llama. Our cache size savings increase linearly with model layer count, ranging from 756-2550 times smaller than the traditional transformer cache for common sizes, enabling inference of extremely large context lengths even on limited hardware. Although autoregressive generation has O(n) time complexity per token because of attention, pre-fill computation of the entire initial cache state for a submitted context costs only O(1) time per token due to the use of a recurrent neural network (RNN) to generate this cache. We release our trained weights and training code under the Apache 2.0 license for community use.

Finetuning large language models (LLMs) has been empirically effective on a variety of downstream tasks. Existing approaches to finetuning an LLM either focus on parameter-efficient finetuning, which only updates a small number of trainable parameters, or attempt to reduce the memory footprint during the training phase of the finetuning. Typically, the memory footprint during finetuning stems from three contributors: model weights, optimizer states, and intermediate activations. However, existing works still require considerable memory and none can simultaneously mitigate memory footprint for all three sources. In this paper, we present Quantized Side Tuing (QST), which enables memory-efficient and fast finetuning of LLMs by operating through a dual-stage process. First, QST quantizes an LLM's model weights into 4-bit to reduce the memory footprint of the LLM's original weights; QST also introduces a side network separated from the LLM, which utilizes the hidden states of the LLM to make task-specific predictions. Using a separate side network avoids performing backpropagation through the LLM, thus reducing the memory requirement of the intermediate activations. Furthermore, QST leverages several low-rank adaptors and gradient-free downsample modules to significantly reduce the trainable parameters, so as to save the memory footprint of the optimizer states. Experiments show that QST can reduce the total memory footprint by up to 2.3 times and speed up the finetuning process by up to 3 times while achieving competent performance compared with the state-of-the-art. When it comes to full finetuning, QST can reduce the total memory footprint up to 7 times.

Nowadays, Large Language Models (LLMs) have been trained using extended context lengths to foster more creative applications. However, long context training poses great challenges considering the constraint of GPU memory. It not only leads to substantial activation memory consumption during training, but also incurs considerable memory fragmentation. To facilitate long context training, existing frameworks have adopted strategies such as recomputation and various forms of parallelisms. Nevertheless, these techniques rely on redundant computation or extensive communication, resulting in low Model FLOPS Utilization (MFU). In this paper, we propose MEMO, a novel LLM training framework designed for fine-grained activation memory management. Given the quadratic scaling of computation and linear scaling of memory with sequence lengths when using FlashAttention, we offload memory-consuming activations to CPU memory after each layer's forward pass and fetch them during the backward pass. To maximize the swapping of activations without hindering computation, and to avoid exhausting limited CPU memory, we implement a token-wise activation recomputation and swapping mechanism. Furthermore, we tackle the memory fragmentation issue by employing a bi-level Mixed Integer Programming (MIP) approach, optimizing the reuse of memory across transformer layers. Empirical results demonstrate that MEMO achieves an average of 2.42x and 2.26x MFU compared to Megatron-LM and DeepSpeed, respectively. This improvement is attributed to MEMO's ability to minimize memory fragmentation, reduce recomputation and intensive communication, and circumvent the delays associated with the memory reorganization process due to fragmentation. By leveraging fine-grained activation memory management, MEMO facilitates efficient training of 7B LLM with 1 million sequence length on just 8 A800 GPUs, achieving an MFU of 52.30%.

Practical large language model (LLM) services may involve a long system prompt, which specifies the instructions, examples, and knowledge documents of the task and is reused across numerous requests. However, the long system prompt causes throughput/latency bottlenecks as the cost of generating the next token grows w.r.t. the sequence length. This paper aims to improve the efficiency of LLM services that involve long system prompts. Our key observation is that handling these system prompts requires heavily redundant memory accesses in existing causal attention computation algorithms. Specifically, for batched requests, the cached hidden states (i.e., key-value pairs) of system prompts are transferred from off-chip DRAM to on-chip SRAM multiple times, each corresponding to an individual request. To eliminate such a redundancy, we propose RelayAttention, an attention algorithm that allows reading these hidden states from DRAM exactly once for a batch of input tokens. RelayAttention is a free lunch: it maintains the generation quality while requiring no model retraining, as it is based on a mathematical reformulation of causal attention.

We present Hydra as an architectural proposal for hybrid long-context language models that combine conditional computation, long-context memory mechanisms, and sparse mixture-of-experts within an approximately 1.6B parameter design envelope. Hydra integrates a Mamba-style Structured State Space Model (SSM) backbone with intermittent sparse global attention, chunk-level MoE feed-forward routing, and dual (workspace plus factual PKM) memories. We formalize the component interfaces, give transparent parameter and complexity accounting, and outline a staged curriculum intended to stably activate the parts. We accompany the specification with illustrative toy-scale prototype measurements (tens of millions of parameters on synthetic data) whose sole purpose is to demonstrate implementation feasibility and qualitative scaling behaviors (for example, long-context throughput crossover and controllable expert routing), not to claim competitive full-scale performance. We explicitly delineate assumptions and open risks (training complexity, memory utilization, specialization dynamics) and position Hydra as a blueprint to stimulate empirical follow-up rather than a finished system. By combining SSM efficiency, selective sparse attention, MoE capacity, and learnable memory, Hydra sketches a path toward modular, input-adaptive long-context language models; validating end-task gains at target scale remains future work.

Memory-based self-evolution has emerged as a promising paradigm for coding agents. However, existing approaches typically restrict memory utilization to homogeneous task domains, failing to leverage the shared infrastructural foundations, such as runtime environments and programming languages, that exist across diverse real-world coding problems. To address this limitation, we investigate Memory Transfer Learning (MTL) by harnessing a unified memory pool from heterogeneous domains. We evaluate performance across 6 coding benchmarks using four memory representations, ranging from concrete traces to abstract insights. Our experiments demonstrate that cross-domain memory improves average performance by 3.7\%, primarily by transferring meta-knowledge, such as validation routines, rather than task-specific code. Importantly, we find that abstraction dictates transferability; high-level insights generalize well, whereas low-level traces often induce negative transfer due to excessive specificity. Furthermore, we show that transfer effectiveness scales with the size of the memory pool, and memory can be transferred even between different models. Our work establishes empirical design principles for expanding memory utilization beyond single-domain silos. Project page: https://memorytransfer.github.io/

AI coding agents operate in a paradox: they possess vast parametric knowledge yet cannot remember a conversation from an hour ago. Existing memory systems store text in vector databases with single-channel retrieval, require cloud LLMs for core operations, and implement none of the cognitive processes that make human memory effective. We present SuperLocalMemory V3.3 ("The Living Brain"), a local-first agent memory system implementing the full cognitive memory taxonomy with mathematical lifecycle dynamics. Building on the information-geometric foundations of V3.2 (arXiv:2603.14588), we introduce five contributions: (1) Fisher-Rao Quantization-Aware Distance (FRQAD) -- a new metric on the Gaussian statistical manifold achieving 100% precision at preferring high-fidelity embeddings over quantized ones (vs 85.6% for cosine), with zero prior art; (2) Ebbinghaus Adaptive Forgetting with lifecycle-aware quantization -- the first mathematical forgetting curve in local agent memory coupled to progressive embedding compression, achieving 6.7x discriminative power; (3) 7-channel cognitive retrieval spanning semantic, keyword, entity graph, temporal, spreading activation, consolidation, and Hopfield associative channels, achieving 70.4% on LoCoMo in zero-LLM Mode A; (4) memory parameterization implementing Long-Term Implicit memory via soft prompts; (5) zero-friction auto-cognitive pipeline automating the complete memory lifecycle. On LoCoMo, V3.3 achieves 70.4% in Mode A (zero-LLM), with +23.8pp on multi-hop and +12.7pp on adversarial. V3.2 achieved 74.8% Mode A and 87.7% Mode C; the 4.4pp gap reflects a deliberate architectural trade-off. SLM V3.3 is open source under the Elastic License 2.0, runs entirely on CPU, with over 5,000 monthly downloads.

Large Language Models (LLMs) have been widely adopted to process long-context tasks. However, the large memory overhead of the key-value (KV) cache poses significant challenges in long-context scenarios. Existing training-free KV cache compression methods typically focus on quantization and token pruning, which have compression limits, and excessive sparsity can lead to severe performance degradation. Other methods design new architectures with less KV overhead but require significant training overhead. To address the above two drawbacks, we further explore the redundancy in the channel dimension and apply an architecture-level design with minor training costs. Therefore, we introduce CSKV, a training-efficient Channel Shrinking technique for KV cache compression: (1) We first analyze the singular value distribution of the KV cache, revealing significant redundancy and compression potential along the channel dimension. Based on this observation, we propose using low-rank decomposition for key and value layers and storing the low-dimension features. (2) To preserve model performance, we introduce a bi-branch KV cache, including a window-based full-precision KV cache and a low-precision compressed KV cache. (3) To reduce the training costs, we minimize the layer-wise reconstruction loss for the compressed KV cache instead of retraining the entire LLMs. Extensive experiments show that CSKV can reduce the memory overhead of the KV cache by 80% while maintaining the model's long-context capability. Moreover, we show that our method can be seamlessly combined with quantization to further reduce the memory overhead, achieving a compression ratio of up to 95%.

External memory systems are pivotal for enabling Large Language Model (LLM) agents to maintain persistent knowledge and perform long-horizon decision-making. Existing paradigms typically follow a two-stage process: computationally expensive memory construction (e.g., structuring data into graphs) followed by naive retrieval-augmented generation. However, our empirical analysis reveals two fundamental limitations: complex construction incurs high costs with marginal performance gains, and simple context concatenation fails to bridge the gap between retrieval recall and reasoning accuracy. To address these challenges, we propose CoM (Chain-of-Memory), a novel framework that advocates for a paradigm shift toward lightweight construction paired with sophisticated utilization. CoM introduces a Chain-of-Memory mechanism that organizes retrieved fragments into coherent inference paths through dynamic evolution, utilizing adaptive truncation to prune irrelevant noise. Extensive experiments on the LongMemEval and LoCoMo benchmarks demonstrate that CoM outperforms strong baselines with accuracy gains of 7.5%-10.4%, while drastically reducing computational overhead to approximately 2.7% of token consumption and 6.0% of latency compared to complex memory architectures.

Recent advances in 1-bit Large Language Models (LLMs), such as BitNet and BitNet b1.58, present a promising approach to enhancing the efficiency of LLMs in terms of speed and energy consumption. These developments also enable local LLM deployment across a broad range of devices. In this work, we introduce bitnet.cpp, a tailored software stack designed to unlock the full potential of 1-bit LLMs. Specifically, we develop a set of kernels to support fast and lossless inference of ternary BitNet b1.58 LLMs on CPUs. Extensive experiments demonstrate that bitnet.cpp achieves significant speedups, ranging from 2.37x to 6.17x on x86 CPUs and from 1.37x to 5.07x on ARM CPUs, across various model sizes. The code is available at https://github.com/microsoft/BitNet.

Large language models (LLMs) have shown remarkable advances in supporting long-context comprehension and processing tasks. However, scaling the generation inference of LLMs to such long contexts incurs significant additional computation load, and demands a substantial GPU memory footprint to maintain the key-value (KV) cache of transformer-based LLMs. Existing KV cache compression methods, such as quantization, face memory bottlenecks as context length increases, while static-sized caches, such as eviction, suffer from inefficient policies. These limitations restrict deployment on consumer-grade devices like a single Nvidia 4090 GPU. To overcome this, we propose Locret, a framework for long-context LLM inference that introduces retaining heads to evaluate the causal importance of KV cache units, allowing for more accurate eviction within a fixed cache size. Locret is fine-tuned on top of the frozen backbone LLM using a minimal amount of data from standard long-context SFT datasets. During inference, we evict low-importance cache units along with a chunked prefill pattern, significantly reducing peak GPU memory usage. We conduct an extensive empirical study to evaluate Locret, where the experimental results show that Locret outperforms the recent competitive approaches, including InfLLM, Quantization, SirLLM, and MInference, in terms of memory efficiency and the quality of generated contents -- Locret achieves over a 20x and 8x KV cache compression ratio compared to the full KV cache for Phi-3-mini-128K and Llama-3.1-8B-instruct. Additionally, Locret can be combined with other methods, such as quantization and token merging. To our knowledge, Locret is the first framework capable of deploying Llama-3.1-8B or similar models on a single Nvidia 4090 GPU, enabling 128K long-context inference without compromising generation quality, and requiring little additional system optimizations.

Large language models have revolutionized data processing in numerous domains, with their ability to handle extended context reasoning receiving notable recognition. To speed up inference, maintaining a key-value (KV) cache memory is essential. Nonetheless, the growing demands for KV cache memory create significant hurdles for efficient implementation. This work introduces a novel architecture, Cross-Layer Latent Attention (CLLA), aimed at compressing the KV cache to less than 2% of its original size while maintaining comparable performance levels. CLLA integrates multiple aspects of KV cache compression, including attention head/dimension reduction, layer sharing, and quantization techniques, into a cohesive framework. Our extensive experiments demonstrate that CLLA achieves lossless performance on most tasks while utilizing minimal KV cache, marking a significant advancement in practical KV cache compression.

In contrast to RNNs, which compress previous tokens into a single hidden state, Transformers can attend to all previous tokens directly. However, standard Transformers only use representations from the immediately preceding layer. In this paper, we show that this design choice causes representation collapse and leads to suboptimal performance. To address this issue, we introduce Layer-Integrated Memory (LIMe), a simple yet powerful approach that preserves the model's overall memory footprint while expanding its representational capacity by allowing access to hidden states from earlier layers. Through extensive experiments across various architectures and different lookup mechanisms, we demonstrate consistent performance improvements on a wide range of tasks. Moreover, our analysis of the learned representation dynamics and our exploration of depthwise circuits reveal how LIMe integrates information across layers, pointing to promising directions for future research.

Long-sequence modeling faces a fundamental trade-off between the efficiency of compressive fixed-size memory in RNN-like models and the fidelity of lossless growing memory in attention-based Transformers. Inspired by the Multi-Store Model in cognitive science, we introduce a memory framework of artificial neural networks. Our method maintains a sliding window of the Transformer's KV cache as lossless short-term memory, while a learnable module termed Artificial Hippocampus Network (AHN) recurrently compresses out-of-window information into a fixed-size compact long-term memory. To validate this framework, we instantiate AHNs using modern RNN-like architectures, including Mamba2, DeltaNet, and Gated DeltaNet. Extensive experiments on long-context benchmarks LV-Eval and InfiniteBench demonstrate that AHN-augmented models consistently outperform sliding window baselines and achieve performance comparable or even superior to full-attention models, while substantially reducing computational and memory requirements. For instance, augmenting the Qwen2.5-3B-Instruct with AHNs reduces inference FLOPs by 40.5% and memory cache by 74.0%, while improving its average score on LV-Eval (128k sequence length) from 4.41 to 5.88. Code is available at: https://github.com/ByteDance-Seed/AHN.

The pervasive "memory wall" bottleneck is significantly amplified in modern large-scale Mixture-of-Experts (MoE) architectures. MoE's inherent architectural sparsity leads to sparse arithmetic compute and also introduces substantial activation memory overheads -- driven by large token routing buffers and the need to materialize and buffer intermediate tensors. This memory pressure limits the maximum batch size and sequence length that can fit on GPUs, and also results in excessive data movements that hinders performance and efficient model scaling. We present MoEBlaze, a memory-efficient MoE training framework that addresses these issues through a co-designed system approach: (i) an end-to-end token dispatch and MoE training method with optimized data structures to eliminate intermediate buffers and activation materializing, and (ii) co-designed kernels with smart activation checkpoint to mitigate memory footprint while simultaneously achieving better performance. We demonstrate that MoEBlaze can achieve over 4x speedups and over 50% memory savings compared to existing MoE frameworks.

Retrieval-Augmented Generation enhances language models by retrieving relevant information from external knowledge bases, relying on high-dimensional vector embeddings typically stored in float32 precision. However, storing these embeddings at scale presents significant memory challenges. To address this issue, we systematically investigate on MTEB benchmark two complementary optimization strategies: quantization, evaluating standard formats (float16, int8, binary) and low-bit floating-point types (float8), and dimensionality reduction, assessing methods like PCA, Kernel PCA, UMAP, Random Projections and Autoencoders. Our results show that float8 quantization achieves a 4x storage reduction with minimal performance degradation (<0.3%), significantly outperforming int8 quantization at the same compression level, being simpler to implement. PCA emerges as the most effective dimensionality reduction technique. Crucially, combining moderate PCA (e.g., retaining 50% dimensions) with float8 quantization offers an excellent trade-off, achieving 8x total compression with less performance impact than using int8 alone (which provides only 4x compression). To facilitate practical application, we propose a methodology based on visualizing the performance-storage trade-off space to identify the optimal configuration that maximizes performance within their specific memory constraints.

Large Language Models (LLMs) have revolutionized a wide range of domains such as natural language processing, computer vision, and multi-modal tasks due to their ability to comprehend context and perform logical reasoning. However, the computational and memory demands of LLMs, particularly during inference, pose significant challenges when scaling them to real-world, long-context, and real-time applications. Key-Value (KV) cache management has emerged as a critical optimization technique for accelerating LLM inference by reducing redundant computations and improving memory utilization. This survey provides a comprehensive overview of KV cache management strategies for LLM acceleration, categorizing them into token-level, model-level, and system-level optimizations. Token-level strategies include KV cache selection, budget allocation, merging, quantization, and low-rank decomposition, while model-level optimizations focus on architectural innovations and attention mechanisms to enhance KV reuse. System-level approaches address memory management, scheduling, and hardware-aware designs to improve efficiency across diverse computing environments. Additionally, the survey provides an overview of both text and multimodal datasets and benchmarks used to evaluate these strategies. By presenting detailed taxonomies and comparative analyses, this work aims to offer useful insights for researchers and practitioners to support the development of efficient and scalable KV cache management techniques, contributing to the practical deployment of LLMs in real-world applications. The curated paper list for KV cache management is in: https://github.com/TreeAI-Lab/Awesome-KV-Cache-Management{https://github.com/TreeAI-Lab/Awesome-KV-Cache-Management}.

Self-evolving memory systems are unprecedentedly reshaping the evolutionary paradigm of large language model (LLM)-based agents. Prior work has predominantly relied on manually engineered memory architectures to store trajectories, distill experience, and synthesize reusable tools, enabling agents to evolve on the fly within environment interactions. However, this paradigm is fundamentally constrained by the staticity of the memory system itself: while memory facilitates agent-level evolving, the underlying memory architecture cannot be meta-adapted to diverse task contexts. To address this gap, we propose MemEvolve, a meta-evolutionary framework that jointly evolves agents' experiential knowledge and their memory architecture, allowing agent systems not only to accumulate experience but also to progressively refine how they learn from it. To ground MemEvolve in prior research and foster openness in future self-evolving systems, we introduce EvolveLab, a unified self-evolving memory codebase that distills twelve representative memory systems into a modular design space (encode, store, retrieve, manage), providing both a standardized implementation substrate and a fair experimental arena. Extensive evaluations on four challenging agentic benchmarks demonstrate that MemEvolve achieves (I) substantial performance gains, improving frameworks such as SmolAgent and Flash-Searcher by up to 17.06%; and (II) strong cross-task and cross-LLM generalization, designing memory architectures that transfer effectively across diverse benchmarks and backbone models.

Transformer-based large language models (LLMs) demonstrate impressive performance in long context generation. Extending the context length has disproportionately shifted the memory footprint of LLMs during inference to the key-value cache (KV cache). In this paper, we propose HEADINFER, which offloads the KV cache to CPU RAM while avoiding the need to fully store the KV cache for any transformer layer on the GPU. HEADINFER employs a fine-grained, head-wise offloading strategy, maintaining only selective attention heads KV cache on the GPU while computing attention output dynamically. Through roofline analysis, we demonstrate that HEADINFER maintains computational efficiency while significantly reducing memory footprint. We evaluate HEADINFER on the Llama-3-8B model with a 1-million-token sequence, reducing the GPU memory footprint of the KV cache from 128 GB to 1 GB and the total GPU memory usage from 207 GB to 17 GB, achieving a 92% reduction compared to BF16 baseline inference. Notably, HEADINFER enables 4-million-token inference with an 8B model on a single consumer GPU with 24GB memory (e.g., NVIDIA RTX 4090) without approximation methods.

Optimizer states are a major source of memory consumption for training neural networks, limiting the maximum trainable model within given memory budget. Compressing the optimizer states from 32-bit floating points to lower bitwidth is promising to reduce the training memory footprint, while the current lowest achievable bitwidth is 8-bit. In this work, we push optimizer states bitwidth down to 4-bit through a detailed empirical analysis of first and second moments. Specifically, we find that moments have complicated outlier patterns, that current block-wise quantization cannot accurately approximate. We use a smaller block size and propose to utilize both row-wise and column-wise information for better quantization. We further identify a zero point problem of quantizing the second moment, and solve this problem with a linear quantizer that excludes the zero point. Our 4-bit optimizers are evaluated on a wide variety of benchmarks including natural language understanding, machine translation, image classification, and instruction tuning. On all the tasks our optimizers can achieve comparable accuracy with their full-precision counterparts, while enjoying better memory efficiency.

A critical approach for efficiently deploying computationally demanding large language models (LLMs) is Key-Value (KV) caching. The KV cache stores key-value states of previously generated tokens, significantly reducing the need for repetitive computations and thereby lowering latency in autoregressive generation. However, the size of the KV cache grows linearly with sequence length, posing challenges for applications requiring long context input and extensive sequence generation. In this paper, we present a simple yet effective approach, called MiniCache, to compress the KV cache across layers from a novel depth perspective, significantly reducing the memory footprint for LLM inference. Our approach is based on the observation that KV cache states exhibit high similarity between the adjacent layers in the middle-to-deep portion of LLMs. To facilitate merging, we propose disentangling the states into the magnitude and direction components, interpolating the directions of the state vectors while preserving their lengths unchanged. Furthermore, we introduce a token retention strategy to keep highly distinct state pairs unmerged, thus preserving the information with minimal additional storage overhead. Our MiniCache is training-free and general, complementing existing KV cache compression strategies, such as quantization and sparsity. We conduct a comprehensive evaluation of MiniCache utilizing various models including LLaMA-2, LLaMA-3, Phi-3, Mistral, and Mixtral across multiple benchmarks, demonstrating its exceptional performance in achieving superior compression ratios and high throughput. On the ShareGPT dataset, LLaMA-2-7B with 4-bit MiniCache achieves a remarkable compression ratio of up to 5.02x, enhances inference throughput by approximately 5x, and reduces the memory footprint by 41% compared to the FP16 full cache baseline, all while maintaining near-lossless performance.

Memory is critical for enabling large language model (LLM) based agents to maintain coherent behavior over long-horizon interactions. However, existing agent memory systems suffer from two key gaps: they rely on a one-size-fits-all memory structure and do not model memory structure selection as a context-adaptive decision, limiting their ability to handle heterogeneous interaction patterns and resulting in suboptimal performance. We propose a unified framework, FluxMem, that enables adaptive memory organization for LLM agents. Our framework equips agents with multiple complementary memory structures. It explicitly learns to select among these structures based on interaction-level features, using offline supervision derived from downstream response quality and memory utilization. To support robust long-horizon memory evolution, we further introduce a three-level memory hierarchy and a Beta Mixture Model-based probabilistic gate for distribution-aware memory fusion, replacing brittle similarity thresholds. Experiments on two long-horizon benchmarks, PERSONAMEM and LoCoMo, demonstrate that our method achieves average improvements of 9.18% and 6.14%.

Recent research, such as BitNet, is paving the way for a new era of 1-bit Large Language Models (LLMs). In this work, we introduce a 1-bit LLM variant, namely BitNet b1.58, in which every single parameter (or weight) of the LLM is ternary {-1, 0, 1}. It matches the full-precision (i.e., FP16 or BF16) Transformer LLM with the same model size and training tokens in terms of both perplexity and end-task performance, while being significantly more cost-effective in terms of latency, memory, throughput, and energy consumption. More profoundly, the 1.58-bit LLM defines a new scaling law and recipe for training new generations of LLMs that are both high-performance and cost-effective. Furthermore, it enables a new computation paradigm and opens the door for designing specific hardware optimized for 1-bit LLMs.

Quantization methods reduce the number of bits required to represent each parameter in a model, trading accuracy for smaller memory footprints and inference latencies. However, the final model size depends on both the number of parameters of the original model and the rate of compression. For example, a 30B 8-bit model and a 60B 4-bit model have the same number of bits but may have very different zero-shot accuracies. In this work, we study this trade-off by developing inference scaling laws of zero-shot performance in Large Language Models (LLMs) to determine the bit-precision and model size that maximizes zero-shot performance. We run more than 35,000 experiments with 16-bit inputs and k-bit parameters to examine which zero-shot quantization methods improve scaling for 3 to 8-bit precision at scales of 19M to 176B parameters across the LLM families BLOOM, OPT, NeoX/Pythia, and GPT-2. We find that it is challenging to improve the bit-level scaling trade-off, with the only improvements being the use of a small block size -- splitting the parameters into small independently quantized blocks -- and the quantization data type being used (e.g., Int vs Float). Overall, our findings show that {4-bit} precision is almost universally optimal for total model bits and zero-shot accuracy.

Large language models (LLMs) have revolutionized AI, but are constrained by limited context windows, hindering their utility in tasks like extended conversations and document analysis. To enable using context beyond limited context windows, we propose virtual context management, a technique drawing inspiration from hierarchical memory systems in traditional operating systems that provide the appearance of large memory resources through data movement between fast and slow memory. Using this technique, we introduce MemGPT (Memory-GPT), a system that intelligently manages different memory tiers in order to effectively provide extended context within the LLM's limited context window, and utilizes interrupts to manage control flow between itself and the user. We evaluate our OS-inspired design in two domains where the limited context windows of modern LLMs severely handicaps their performance: document analysis, where MemGPT is able to analyze large documents that far exceed the underlying LLM's context window, and multi-session chat, where MemGPT can create conversational agents that remember, reflect, and evolve dynamically through long-term interactions with their users. We release MemGPT code and data for our experiments at https://memgpt.ai.

The rapid growth of large language models (LLMs) has outpaced the evolution of single-GPU hardware, making model scale increasingly constrained by memory capacity rather than computation. While modern training systems extend GPU memory through distributed parallelism and offloading across CPU and storage tiers, they fundamentally retain a GPU-centric execution paradigm in which GPUs host persistent model replicas and full autograd graphs. As a result, scaling large models remains tightly coupled to multi-GPU clusters, complex distributed runtimes, and unpredictable host memory consumption, creating substantial barriers for node-scale post-training workloads such as instruction tuning, alignment, and domain adaptation. We present Horizon-LM, a memory-centric training system that redefines the roles of CPU and GPU for large-model optimization. Horizon-LM treats host memory as the authoritative parameter store and uses GPUs solely as transient compute engines through a CPU-master, GPU-template execution model. By eliminating persistent GPU-resident modules and autograd graphs, employing explicit recomputation with manual gradient propagation, and introducing a pipelined double-buffered execution engine, Horizon-LM decouples model scale from GPU count and bounds memory usage to the theoretical parameter footprint. On a single H200 GPU with 1.5\,TB host RAM, Horizon-LM reliably trains models up to 120B parameters. On a standard single A100 machine, Horizon-LM achieves up to 12.2times higher training throughput than DeepSpeed ZeRO-3 with CPU offloading while preserving numerical correctness. Across platforms and scales, Horizon-LM sustains high device utilization and predictable memory growth, demonstrating that host memory, not GPU memory, defines the true feasibility boundary for node-scale large-model training.