On-Device Language Model Inference

Ask about this topic

Overview

The field studies how to run useful language models locally, where limited memory, battery life, heat, and compute can constrain both model size and response speed. Common approaches include quantization and pruning, compact or mixture-of-experts models, KV-cache compression, speculative decoding, and hardware-aware runtimes for mobile CPUs, GPUs, and NPUs. Other systems split work between a device and the cloud or route tasks among local models to balance privacy, cost, and capability. A central tension is that gains in speed or memory do not always preserve accuracy, reliability, or energy efficiency, so evaluation increasingly measures full workloads on real devices rather than relying on model size or peak throughput alone.

Latest papers

CardsList

170 papers26 papers in the last four weeks, down 10% on the four weeks before. 0.2% of all new papers.

May 16, 2026

OpenJarvis: Personal AI, On Personal Devices

Personal AI stacks, like OpenClaw and Hermes Agent, are becoming central to daily work, yet they route nearly every query (often over sensitive local data) to cloud-hosted frontier models. Replacing frontier models with local models inside existing stacks does not work: swapping Claude Opus 4.6 for Qwen3.5-9B drops accuracy by 25-39 pp across personal AI tasks like PinchBench and GAIA. Existing stacks bundle agentic prompts, tool descriptions, memory configuration, and runtime settings around a specific cloud model. Only the prompts can be tuned, and state-of-the-art prompt optimizers close just 5 pp of the local-cloud gap on their own. This motivates a decomposed personal AI stack: one that exposes individual primitives which can be optimized individually or jointly to close the local-cloud gap. We present OpenJarvis, an architecture that represents a personal AI system as a typed spec over five primitives: Intelligence, Engine, Agents, Tools & Memory, and Learning. Each primitive is an independently editable field, making the stack end-to-end optimizable and measurable against accuracy, cost, and latency. Towards closing the local-cloud gap without surrendering local-model properties, OpenJarvis introduces LLM-guided spec search, a local-cloud collaboration in which frontier cloud models propose edits across the spec at search time, only non-regressing edits are accepted, and the resulting spec runs entirely on-device at inference time. With LLM-guided spec search, on-device specs match or exceed cloud accuracy on 4 of 8 benchmarks and land within 3.2 pp of the best cloud baseline on average. They also reduce marginal API cost by ~800x and end-to-end latency by 4x.

Jon Saad-Falcon, Avanika Narayan, Robby Manihani +10

Stanford University · Lambda Labs

On-Device Language Model InferenceLLM AgentsPersonal AI Agents
AskPDF
May 16, 2026

Lever: Speculative LLM Inference on Smartphones

Large language models (LLMs) are increasingly needed for interactive mobile applications, but high-quality models exceed the limited DRAM available on smartphones. Flash storage can hold larger models, yet flash-backed inference is slow because autoregressive decoding repeatedly invokes the target model and incurs costly I/O. We observe that speculative decoding is a natural fit for this setting: a small draft model can remain in DRAM, while a larger flash-resident target model verifies multiple candidate tokens per invocation. However, existing methods assume server-class accelerators and fail to account for prolonged I/O latency, limited computation parallelism, and irregular speculation execution. We present Lever, an end-to-end system for efficient flash-backed LLM inference on smartphones. Lever jointly optimizes the three stages of speculative decoding under mobile constraints. For drafting, it builds token trees using an I/O- and compute-aware gain-cost objective. For verification, it prunes low-value branches through early-exit prediction to reduce target-model computation. For execution, it maps speculation efficiently across mobile CPU-NPU hardware to improve utilization. Comprehensive evaluations show that Lever reduces inference latency by an average of 2.93x over baseline flash-offloaded inference and 1.50x over conventional speculative decoding, narrowing the latency gap between flash-backed and memory-resident LLM inference.

Tuowei Wang, Fengzu Li, Yanfan Sun +2

Tsinghua University · Beijing, China · Beihang University +1

On-Device Language Model InferenceLLM InferenceSpeculative DecodingLLM Inference Acceleration
AskPDF
May 12, 2026

CR^2: Cost-Aware Risk-Controlled Routing for Wireless Device-Edge LLM Inference

As large language models (LLMs) move from centralized clouds to mobile edge environments, efficient serving must balance latency, energy consumption, and accuracy under constrained device-edge resources. Query-level routing between lightweight on-device models and stronger edge models provides a flexible mechanism to navigate this trade-off. However, existing routers are designed for centralized cloud settings and optimize token-level costs, failing to capture the dynamic latency and energy overheads in wireless edge deployments. In this paper, we formulate mobile edge LLM routing as a deployment-constrained, cost-aware decision problem, and propose CR^2, a two-stage device-edge routing framework. CR^2 decouples a lightweight on-device margin gate from an edge-side utility selector for deferred queries. The margin gate operates on frozen query embeddings and a user-specified cost weight to predict whether local execution is utility-optimal relative to the best edge alternative under the target operating point. We further introduce a conformal risk control (CRC) calibration procedure that maps each operating point to an acceptance threshold, enabling explicit control of the marginal false-acceptance risk under the full-information utility reference. Experiments on the routing task show that CR^2 closely matches a full-information reference router using only device-side signals before deferral. Compared with strong query-level baselines, CR^2 consistently improves the deployable accuracy-cost Pareto frontier and reduces normalized deployment cost by up to 16.9% at matched accuracy.

Nan Xue, Shengkang Chen, Zhiyong Chen +4

Cooperative Medianet Innovation Center, Shanghai Jiao Tong University, Shanghai 200240, China · Department of Broadband Communication, Pengcheng Laboratory, Shenzhen 518000, China · Future Network of Intelligent Institute (FNii), the Chinese University of Hong Kong (Shenzhen), Shenzhen 518172, China +1

On-Device Language Model InferenceLLM InferenceCost-Aware InferenceConformal Risk ControlLLM Routing
AskPDF
May 11, 2026

CATS: Cascaded Adaptive Tree Speculation for Memory-Limited LLM Inference Acceleration

Auto-regressive decoding in Large Language Models (LLMs) is inherently memory-bound: every generation step requires loading the model weights and intermediate results from memory (e.g., High-Bandwidth Memory (HBM) for GPU servers), making throughput bottlenecked by memory bandwidth rather than compute. Speculative decoding addresses this by enabling parallel verification of multiple draft tokens, effectively amortizing the cost of each target-model call. However, existing speculative decoding methods are designed under the assumption that HBM is sufficiently large to hold both the target model and an auxiliary draft model simultaneously -- an assumption that breaks down on memory-constrained devices such as edge platforms with limited DRAM. We analyze the inference bottleneck in this memory-limited regime and propose CATS, a self-speculative decoding framework that conducts cascaded verification and correction based on the memory budget and parameter offloading patterns on memory-limited devices. This design maximizes token acceptance rate and end-to-end speedup while keeping the peak memory footprint on the device equal to that of the target model alone. We evaluate CATS on different models across five benchmarks on real edge devices. CATS can achieve a wall-clock speedup of up to 5.08x with no degradation in generation quality, outperforming the SOTA method by up to 1.45x under edge memory constraints.

Yuning Han, Yangchenchen Jin, Dylan Zhao +1

University of Florida Gainesville, FL 32611

On-Device Language Model InferenceMemory-Efficient InferenceSpeculative DecodingLLM Inference Acceleration
AskPDF
May 11, 2026

Agent-X: Full Pipeline Acceleration of On-device AI Agents

LLM-based agents deliver state-of-the-art performance across tasks but incur high end-to-end latency on edge devices. We introduce Agent-X, a software-only, accuracy-preserving framework that accelerates both the prefill and decode stages of on-device agent workloads. Agent-X's two key components rewrite prompts to leverage prefix caching tailored to agent-specific input-token patterns and enable LLM-free speculative decoding for fast token generation with minimal overhead. On representative agentic workloads, Agent-X achieves a 1.61x end-to-end speedup in real systems with no accuracy loss and can be seamlessly integrated into existing on-device AI agents. To the best of our knowledge, ours is the first to systematically characterize and eliminate latency bottlenecks in on-device agents.

Jinha Chung, Byeongjun Shin, Jiin Kim +1

KAIST Republic of Korea, Daejeon

LLM Inference EfficiencyOn-Device Language Model InferenceLLM Agent Workflow OptimizationPrefix CachingLLM Agents
AskPDF
May 11, 2026

GELATO: Generative Entropy- and Lyapunov-based Adaptive Token Offloading for Device-Edge Speculative LLM Inference

The recent growth of on-device Large Language Model (LLM) inference has driven significant interest in device-edge collaborative LLM inference. As a promising architecture, Speculative Decoding (SD) is increasingly adopted where a lightweight draft model rapidly generates candidate tokens to be verified by a powerful target model. However, a fundamental challenge lies in achieving per-token resource scheduling to effectively adapt SD paradigm to resource-constrained edge environment. This paper proposes a Generative Entropy- and Lyapunov-based Adaptive Token Offloading framework, named GELATO, to maximize decoding throughput under energy constraints in a device-edge collaborative SD system. Specifically, an outer drift-plus-penalty loop makes online decisions to establish a reference drafting budget, managing long-term energy-throughput trade-off. Further, a nested entropy-driven generation mechanism executes early exiting to adapt to per-token dynamic generative uncertainty. Theoretical analysis establishes a rigorous performance bound on long-term throughput for GELATO. Extensive evaluations demonstrate that GELATO achieves a globally optimal tradeoff, outperforming state-of-the-art distributed SD architectures by 64.98% in token throughput and reducing energy consumption by 47.47% under resource-constrained environments, while preserving LLM decoding quality.

Zengzipeng Tang, Yuxuan Sun, Wei Chen +2

School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing 100044, China

On-Device Language Model InferenceEdge ComputingSpeculative DecodingLLM Inference Acceleration
AskPDF
May 9, 2026

Large Language Models over Networks: Collaborative Intelligence under Resource Constraints

Large language models (LLMs) are transforming society, powering applications from smartphone assistants to autonomous driving. Yet cloud-based LLM services alone cannot serve a growing class of applications, including those operating under intermittent connectivity, sub-second latency budgets, data-residency constraints, or sustained high-volume inference. On-device deployment is in turn constrained by limited computation and memory. No single endpoint can deliver high-quality service across this spectrum. This article focuses on collaborative intelligence, a paradigm in which multiple independent LLMs distributed across device and cloud endpoints collaborate at the task level through natural language or structured messages. Such collaboration strives for superior response quality under heterogeneous resource constraints spanning computation, memory, communication, and cost across network tiers. We present collaborative inference along two complementary and composable dimensions: vertical device-cloud collaboration and horizontal multi-agent collaboration, which can be combined into hybrid topologies in practice. We then examine learning to collaborate, addressing the training of routing policies and the development of cooperative capabilities among LLMs. Finally, we identify open research challenges including scaling under resource heterogeneity and trustworthy collaborative intelligence.

Liangqi Yuan, Wenzhi Fang, Shiqiang Wang +2

Purdue University, IN, USA. · University of Exeter, UK. · Princeton University, USA.

LLM Inference EfficiencyOn-Device Language Model InferenceLLM InferenceLLM RoutingMulti-LLM CollaborationEdge-Cloud Collaborative Inference
AskPDF
May 4, 2026

Benchmarking Local Language Models for Social Robots using Edge Devices

Social-educational robots designed for socially interactive pedagogical support, such as the Robot Study Companion (RSC), rely on responsive, privacy-preserving interaction despite severely limited compute. However, there is a gap in systematic benchmarking of language models for edge computing in pedagogical applications. This paper benchmarks 25 open-source language models for local deployment on edge hardware. We evaluate each model across three dimensions: inference efficiency (tokens per second, energy consumption), general knowledge (a six-category MMLU subset), and teaching effectiveness (LLM-rated pedagogical quality), validated against five independent human raters using the Raspberry Pi(RPi)4 as the primary platform, with additional comparisons on the RPi5 and a laptop GPU. Results reveal pronounced trade-offs: throughput and energy efficiency vary by over an order of magnitude across models, MMLU accuracy ranges from near-random to 57.2%, and teaching effectiveness does not correlate monotonically with either metric. Among the evaluated models, Granite4 Tiny Hybrid (7B) achieves a strong overall balance, reaching 2.5 tokens per second, 0.90 tokens per joule, and 54.6% MMLU accuracy; high MMLU accuracy does not appear necessary for strong teaching scores. Human validation on four representative models preserved the automated rank ordering (Pearson r = 0.967, n = 4). Based on these findings, we propose a three-tier local inference architecture for the RSC that balances responsiveness and accuracy on resource-constrained hardware.

Dorian Lamouille, Matevž B. Zorec, Farnaz Baksh +1

University of Tartu Tartu, Estonia · Institute of Computer Science University of Tartu Tartu, Estonia · Institute of Technology University of Tartu Tartu, Estonia

LLM EvaluationOn-Device Language Model InferenceAI in EducationSmall Language ModelsRoboticsSocial Robotics
AskPDF
May 2, 2026

Weight Pruning Amplifies Bias: A Multi-Method Study of Compressed LLMs for Edge AI

Weight pruning is widely advocated for deploying Large Language Models on resource-constrained IoT and edge devices, yet its impact on model fairness remains poorly understood. We conduct a controlled empirical study of three instruction-tuned models (Gemma-2-9b-it, Mistral-7B-Instruct-v0.3, Phi-3.5-mini-instruct) across three pruning methods (Random, Magnitude, Wanda) at four sparsity levels (10-70%) on 12,148 BBQ bias benchmark items with 5 random seeds, totaling 2,368,860 inference records. Our results reveal a Smart Pruning Paradox: activation-aware pruning (Wanda) preserves perplexity nearly perfectly (just 3.5% increase at 50% sparsity for Mistral-7B), yet produces the highest bias amplification, with Stereotype Reliance Score increasing 83.7% and 47-59% of previously unbiased items developing new stereotypical behaviors at 70% sparsity. Random pruning destroys language capability entirely (perplexity exceeding 10410^4 and reaching 10810^8) but produces only random-chance bias. We further show that unstructured pruning provides zero storage savings and zero inference latency reduction on real edge hardware, undermining the primary motivation for its use in IoT deployment. Of 180 dense-vs-pruned comparisons, 141 (78.3%) are significant (p<0.05p < 0.05) with mean ∣h∣=0.305|h| = 0.305. Published quantization studies report up to 21% of responses flipping between biased and unbiased states; our pruning results show transition rates nearly three times higher (47-59%), suggesting pruning poses a categorically greater risk to alignment than quantization. These findings demonstrate that perplexity-based evaluation provides false assurance of behavioral equivalence, and that IoT deployment pipelines require bias-aware validation before deploying pruned models at the edge.

Plawan Kumar Rath, Rahul Maliakkal

Meta

LLM PruningOn-Device Language Model InferenceSocial Bias in Language ModelsLanguage Model Bias Evaluation
AskPDF
May 1, 2026

AgentStop: Terminating Local AI Agents Early to Save Energy in Consumer Devices

Autonomous agents powered by large language models (LLMs) are increasingly used to automate complex, multi-step tasks such as coding or web-based question answering. While remote, cloud-based agents offer scalability and ease of deployment, they raise privacy concerns, depend on network connectivity, and incur recurring API costs. Deploying agents locally on user devices mitigates these issues by preserving data privacy and eliminating usage-based fees. However, agentic workflows are far more resource-intensive than typical LLM interactions. Iterative reasoning, tool use, and failure retries substantially increase token consumption, often expending significant compute without successfully completing tasks. In this work, we investigate the time, token, and energy overhead of locally deployed LLM-based agents on consumer hardware. Our measurements show that agentic execution increases GPU power draw, temperature, and battery drain compared to single-inference workloads. To address this inefficiency, we introduce AgentStop, a lightweight efficiency supervisor that predicts and preemptively terminates trajectories unlikely to succeed. Leveraging low-cost execution signals, such as token-level log probabilities, AgentStop can reduce wasted energy by 15-20% with minimal impact on task performance (<5% utility drop) for challenging web-based question answering and coding benchmarks. These findings position predictive early termination as a practical mechanism for enabling sustainable, privacy-preserving LLM agents on user devices. Our project code and data are available at https://github.com/brave-experiments/AgentStop.

Dzung Pham, Kleomenis Katevas, Ali Shahin Shamsabadi +1

University of Massachusetts Amherst Amherst, Massachusetts, USA · Brave Software London, UK · Brave Software, Imperial College London London, UK

LLM Inference EfficiencyOn-Device Language Model InferenceLLM Agents
AskPDF
May 1, 2026

Silicon Showdown: Performance, Efficiency, and Ecosystem Barriers in Consumer-Grade LLM Inference

The operational landscape of local Large Language Model (LLM) inference has shifted from lightweight models to datacenter-class weights exceeding 70B parameters, creating profound systems challenges for consumer hardware. This paper presents a systematic empirical analysis of the Nvidia and Apple Silicon ecosystems, specifically characterizing the distinct intra-architecture trade-offs required to deploy these massive models. On the Nvidia Blackwell architecture, we identify a critical "Backend Dichotomy" within the TensorRT-LLM stack: while the new NVFP4 quantization format delivers a 1.6x throughput advantage over optimized BF16 baselines (151 tokens/s vs. 92 tokens/s), realizing this performance requires navigating complex runtime constraints that trade startup latency for generation speed. Furthermore, we characterize the "VRAM Wall" for 70B+ models: on discrete GPUs, users face a destructive choice between aggressive quantization (e.g., Q2) that degrades model intelligence to fit in VRAM, or PCIe-bottlenecked CPU offloading, which reduces throughput by over 90% compared to full-GPU execution. Conversely, Apple's Unified Memory Architecture (UMA) circumvents these bottlenecks, enabling linear scaling for 80B parameter models at practical 4-bit precisions. This architectural divergence extends to operational sustainability, where Apple's SoC design demonstrates up to a 23x advantage in energy efficiency (tokens/joule). We conclude that for consumer-grade inference, the optimal hardware is defined by a complex interplay between compute density (Nvidia) and memory capacity (Apple), moderated by the significant "ecosystem friction" of proprietary quantization workflows.

Abdurrahman Javat, Allan Kazakov

Bahcesehir University, Istanbul, Türkiye

LLM Inference EfficiencyLLM QuantizationOn-Device Language Model InferenceEnergy-Efficient ML
AskPDF
Apr 29, 2026

DUAL-BLADE: Dual-Path NVMe-Direct KV-Cache Offloading for Edge LLM Inference

The increasing deployment of Large Language Model (LLM) inference on edge AI systems demands efficient execution under tight memory budgets. A key challenge arises from Key-Value (KV) caches, which often exceed available device memory. Although NVMe-based offloading offers scalable capacity, existing file-based designs rely heavily on the kernel page cache, leading to cache thrashing, unpredictable latency, and high software overhead under memory pressure. We present DUAL-BLADE, a dual-path KV residency framework that dynamically assigns KV tensors to either a page-cache path or an NVMe-direct path based on runtime memory availability. The NVMe-direct path bypasses the filesystem by mapping KV tensors to contiguous logical block address (LBA) regions, enabling low-overhead direct storage access. DUAL-BLADE further incorporates adaptive pipeline parallelism to overlap storage I/O with GPU DMA, improving inference throughput. Our evaluation shows that DUAL-BLADE substantially mitigates I/O bottlenecks, reducing prefill and decode latency by up to 33.1% and 42.4%, respectively, while improving SSD utilization by 2.2x across diverse memory budgets.

Bodon Jeong, Hongsu Byun, Youngjae Kim +4

1Sogang University, Seoul, Republic of Korea · 2Florida State University, FL, USA · 3Samsung Electronics Co.

On-Device Language Model InferenceKV-Cache OffloadingLLM InferenceKV-Cache ManagementMemory-Efficient Inference
AskPDF
Apr 28, 2026

AHASD: Asynchronous Heterogeneous Architecture for LLM Adaptive Drafting Speculative Decoding on Mobile Devices

Speculative decoding enhances the inference efficiency of large language models (LLMs) by generating drafts using a small draft language model (DLM) and verifying them in batches with a large target language model (TLM). However, adaptive drafting inference on a mobile single-NPU-PIM system faces idle overhead in traditional operator-level synchronous execution and wasted computation in asynchronous execution due to fluctuations in draft length. This paper introduces AHASD, a task-level asynchronous mobile NPU-PIM heterogeneous architecture for speculative decoding. Notably, AHASD achieves parallel drafting on the PIM and verification on a single NPU through task-level DLM-TLM decoupling and specifically, it incorporates Entropy-History-Aware Drafting Control and Time-Aware Pre-Verification Control to dynamically manage adaptive drafting algorithm execution and pre-verification timing, suppressing invalid drafting based on low-confidence drafts. Additionally, AHASD integrates Attention Algorithm Units and Gated Task Scheduling Units within LPDDR5-PIM to enable attention link localization and sub-microsecond task switching on the PIM side. Experimental results for different LLMs and adaptive drafting algorithms show that AHASD achieves up to 4.2×\times in throughput and 5.6×\times in energy efficiency improvements over a GPU-only baseline, and 1.5×\times in throughput and 1.24×\times in energy efficiency gains over the state-of-the-art GPU+PIM baseline, with hardware overhead below 3% of the DRAM area.

Ma Zirui, Fan Zhihua, Li Wenxing +4

State Key Lab of Processors, Institute of Computing Technology, Chinese Academy of Sciences · University of Chinese Academy of Sciences

Compute-in-MemoryOn-Device Language Model InferenceSpeculative DecodingLLM Inference Acceleration
AskPDF
Apr 27, 2026

Less is More: Lightweight Prompt Compression for Question Answering Applications on Edge Devices

In agent-driven question answering (QA) applications, retrieval-augmented generation (RAG) is commonly introduced to enhance the response accuracy of large language models (LLMs) by providing additional context. Due to the inherent noise in retrieval results and the coarse granularity of document-level retrieval, the retrieved context often contains substantial redundant information. In this setting, the agent prompt, consisting of the user query and the associated retrieved context, leads to unnecessary computational overhead during LLM inference. Existing prompt compression methods typically rely on auxiliary small language models (SLMs) to estimate context importance. However, such approaches introduce significant memory and computational overhead, which limits their deployment on resource-constrained edge devices. In this paper, we propose CORE, a two-stage sentence-level prompt compression method that eliminates the need for SLMs. In the first stage, CORE constructs an answer set via named entity recognition (NER) and a clue set via semantic matching. In the second stage, CORE refines the clue set using an orthogonal residual retrieval strategy and designs a spatial proximity-based metric to filter the answer set. The two sets are then combined to form the final compressed context. We implement CORE on an NVIDIA Jetson AGX Orin edge device and a Huawei Nova smartphone. Experimental results demonstrate that within a 2000-token budget, CORE improves accuracy by at least 30.19% compared to state-of-the-art baselines, while reducing memory usage by at least 50.47% and achieving at least 1.94 times speedup on the edge device. Moreover, compared to the state-of-the-art LLMLingua2 method, CORE achieves a substantial energy reduction of 95.74% on the smartphone, highlighting its practicality and generalizability for mobile deployments.

Zihuai Xu, Ruofei Hou, Yang Xu +3

Suzhou Institute for Advanced Research, University of Science and Technology of China, Suzhou, Jiangsu, China, 215123

Retrieval-Augmented GenerationOn-Device Language Model InferenceQuestion AnsweringLLM Inference AccelerationPrompt Compression
AskPDF
Apr 27, 2026

Less Is More: Engineering Challenges of On-Device Small Language Model Integration in a Mobile Application

On-device Small Language Models (SLMs) promise fully offline, private AI experiences for mobile users (no cloud dependency, no data leaving the device). But is this promise achievable in practice? This paper presents a longitudinal practitioner case study documenting the engineering challenges of integrating SLMs (Gemma 4 E2B, 2.6B parameters; Qwen3 0.6B, 600M parameters) into Palabrita, a production Android word-guessing game. Over a 5-day development sprint comprising 204 commits (~90 directly AI-related), the system underwent a radical transformation: from an ambitious design where the LLM generated complete structured puzzles (word, category, difficulty, and five hints as JSON) to a pragmatic architecture where curated word lists provide the words and the LLM generates only three short hints, with a deterministic fallback if it fails. We identify five categories of failures specific to on-device SLM integration: output format violations, constraint violations, context quality degradation, latency incompatibility, and model selection instability. For each failure category, we document the observed symptoms, root causes, and the prompt engineering and architectural strategies that effectively mitigated them, including multi-layer defensive parsing, contextual retry with failure feedback, session rotation, progressive prompt hardening, and systematic responsibility reduction. Our findings demonstrate that on-device SLMs are viable for production mobile applications, but only when the developer accepts a fundamental constraint: the most reliable on-device LLM feature is one where the LLM does the least. We distill our experience into eight actionable design heuristics for practitioners integrating SLMs into mobile apps.

William Oliveira

Independent Researcher

On-Device Language Model InferenceSmall Language ModelsLanguage Model Robustness
AskPDF
Apr 27, 2026

BitRL: Reinforcement Learning with 1-bit Quantized Language Models for Resource-Constrained Edge Deployment

The deployment of intelligent reinforcement learning (RL) agents on resource-constrained edge devices remains a fundamental challenge due to the substantial memory, computational, and energy requirements of modern deep learning systems. While large language models (LLMs) have emerged as powerful architectures for decision-making agents, their multi-billion parameter scale confines them to cloud-based deployment, raising concerns about latency, privacy, and connectivity dependence. We introduce BitRL, a framework for building RL agents using 1-bit quantized language models that enables practical on-device learning and inference under severe resource constraints. Leveraging the BitNet b1.58 architecture with ternary weights (-1, 0, +1) and an optimized inference stack, BitRL achieves 10-16x memory reduction and 3-5x energy efficiency improvements over full-precision baselines while maintaining 85-98 percent of task performance across benchmarks. We provide theoretical analysis of quantization as structured parameter perturbation, derive convergence bounds for quantized policy gradients under frozen-backbone architectures, and identify the exploration-stability trade-off in extreme quantization. Our framework systematically integrates 1-bit quantized language models with reinforcement learning for edge deployment and demonstrates effectiveness on commodity hardware.

Md. Ashiq Ul Islam Sajid, Mohammad Sakib Mahmood, Md. Tareq Hasan +3

Department of Computer Science BRAC University Dhaka 1212, Bangladesh · Department of Computer Science Missouri State University Springfield, MO, USA · Department of Computer Science & Engineering Prime University Dhaka, Bangladesh +2

LLM QuantizationRL for Language ModelsOn-Device Language Model InferenceEdge MLLow-Bit Quantization
AskPDF
Apr 25, 2026

An Empirical Evaluation of Locally Deployed LLMs for Bug Detection in Python Code

Large language models (LLMs) have demonstrated strong performance on a wide range of software engineering tasks, including code generation and analysis. However, most prior work relies on cloud-based models or specialized hardware, limiting practical applicability in privacy-sensitive or resource-constrained environments. In this paper, we present a systematic empirical evaluation of two locally deployed LLMs, LLaMA 3.2 and Mistral, for real-world Python bug detection using the BugsInPy benchmark. We evaluate 349 bugs across 17 projects using a zero-shot prompting approach at the function level and an automated keyword-based evaluation framework. Our results show that locally executed models achieve accuracy between 43% and 45%, while producing a large proportion of partially correct responses that identify problematic code regions without pinpointing the exact fix. Performance varies significantly across projects, highlighting the importance of codebase characteristics. The results demonstrate that local models can identify a meaningful share of bugs, though precise localization remains difficult for locally executed LLMs, particularly when handling complex and context dependent bugs in realistic development scenarios.

Jelena Ilić Vulićević

Independent Researcher

LLM EvaluationOn-Device Language Model Inference
AskPDF
Apr 24, 2026

Cloud to Edge: Benchmarking LLM Inference On Hardware-Accelerated Single-Board Computers

Large language models (LLMs) are becoming increasingly capable at small parameter scales. At the same time, conventional cloud-centric deployment introduces challenges around data privacy, latency, and cost that are acute in operational technology and defence environments. Advances in model distillation, quantisation, and affordable edge accelerators now make local LLM inference on single-board computers feasible, but the high dimensionality of the configuration space makes identifying optimal deployments difficult without structured evaluation. Existing LLM-specific edge benchmarking efforts rely on CPU-only inference, poor coverage of genuine single-board computers, and generic evaluation tasks that lack multi-dimensional assessment of hardware effectiveness. This paper proposes a multi-dimensional benchmarking methodology that jointly evaluates inference performance and hardware efficiency across four IoT-suitable edge platform configurations testing single-board computers with the latest available hardware accelerators. Our results reveal the benefits of using hardware accelerators such as NPUs and GPUs, along with multi-dimensional evaluations quantifying the trade-offs between power efficiency, physical device size and token throughput; offering practical guidance for deploying generative AI in privacy-sensitive and connectivity-limited environments such as unmanned vehicles and portable, ruggedised operations.

Harri Renney, Fouad Trad, Michael Mattarock +1

Kaze Technologies Kaze Consulting Bath, BA1 2HN, UK · Electrical and Computer Engineering Lebanese American University Byblos, Lebanon · Carnegie Mellon University Pittsburgh Pennsylvania USA +1

AI Accelerator InferenceOn-Device Language Model InferenceLLM InferenceEdge Computing
AskPDF
Apr 23, 2026

Multilinguality at the Edge: Developing Language Models for the Global South

Where and how language models (LMs) are deployed determines who can benefit from them. However, there are several challenges that prevent effective deployment of LMs in non-English-speaking and hardware constrained communities in the Global South. We call this challenge the last mile: the intersection of multilinguality and edge deployment, where the goals are aligned but the technical requirements often compete. Studying these two fields together is both a need, as linguistically diverse communities often face the most severe infrastructure constraints, and an opportunity, as edge and multilingual NLP research remain largely siloed. To understand the state of the art and the challenges of combining the two areas, we survey 232 papers that tackle this problem across the language modelling pipeline, from data collection to development and deployment. We also discuss open questions and provide actionable recommendations for different stakeholders in the NLP ecosystem. Finally, we hope that this work contributes to the development of inclusive and equitable language technologies.

Lester James V. Miranda, Songbo Hu, Roi Reichart +1

Language Technology Lab, University of Cambridge, United Kingdom · Technion – Israel Institute of Technology, Israel

Multilingual Language ModelsOn-Device Language Model InferenceLow-Resource Language Processing
AskPDF
Apr 23, 2026

SparKV: Overhead-Aware KV Cache Loading for Efficient On-Device LLM Inference

Efficient inference for on-device Large Language Models (LLMs) remains challenging due to limited hardware resources and the high cost of the prefill stage, which processes the full input context to construct Key-Value (KV) caches. We present SparKV, an adaptive KV loading framework that combines cloud-based KV streaming with on-device computation. SparKV models the cost of individual KV chunks and decides whether each chunk should be streamed or computed locally, while overlapping the two execution paths to reduce latency. To handle fluctuations in wireless connectivity and edge resource availability, SparKV further refines offline-generated schedules at runtime to rebalance communication and computation costs. Experiments across diverse datasets, LLMs, and edge devices show that SparKV reduces Time-to-First-Token by 1.3$x-5.1x with negligible impact on response quality, while lowering per-request energy consumption by 1.5x to 3.3x, demonstrating its robustness and practicality for real-world on-device deployment.

Hongyao Liu, Liuqun Zhai, Junyi Wang +3

Department of Computer Science, City University of Hong Kong, Hong Kong SAR.

On-Device Language Model InferenceKV-Cache OffloadingEdge-Cloud Collaborative InferenceKV-Cache ManagementLLM Inference Acceleration
AskPDF
Apr 21, 2026

Micro Language Models Enable Instant Responses

Edge devices such as smartwatches and smart glasses cannot continuously run even the smallest 100M-1B parameter language models due to power and compute constraints, yet cloud inference introduces multi-second latencies that break the illusion of a responsive assistant. We introduce micro language models (μμLMs): ultra-compact models (8M-30M parameters) that instantly generate the first 4-8 words of a contextually grounded response on-device, while a cloud model completes it; thus, masking the cloud latency. We show that useful language generation survives at this extreme scale with our models matching several 70M-256M-class existing models. We design a collaborative generation framework that reframes the cloud model as a continuator rather than a respondent, achieving seamless mid-sentence handoffs and structured graceful recovery via three error correction methods when the local opener goes wrong. Empirical results show that μμLMs can initiate responses that larger models complete seamlessly, demonstrating that orders-of-magnitude asymmetric collaboration is achievable and unlocking responsive AI for extremely resource-constrained devices. The model checkpoint and demo are available at https://github.com/Sensente/micro_language_model_swen_project.

Wen Cheng, Tuochao Chen, Karim Helwani +3

Paul G. Allen School of Computer Science & Engineering, University of Washington · Meta AI

On-Device Language Model InferenceSmall Language ModelsLLM Inference AccelerationMulti-LLM Collaboration
AskPDF
Apr 20, 2026

Toward Zero-Egress Psychiatric AI: On-Device LLM Deployment for Privacy-Preserving Mental Health Decision Support

Privacy represents one of the most critical yet underaddressed barriers to AI adoption in mental healthcare -- particularly in high-sensitivity operational environments such as military, correctional, and remote healthcare settings, where the risk of patient data exposure can deter help-seeking behavior entirely. Existing AI-enabled psychiatric decision support systems predominantly rely on cloud-based inference pipelines, requiring sensitive patient data to leave the device and traverse external servers, creating unacceptable privacy and security risks in these contexts. In this paper, we propose a zero-egress, on-device AI platform for privacy-preserving psychiatric decision support, deployed as a cross-platform mobile application. The proposed system extends our prior work on fine-tuned LLM consortiums for psychiatric diagnosis standardization by fundamentally re-architecting the inference pipeline for fully local execution -- ensuring that no patient data is transmitted to, processed by, or stored on any external server at any stage. The platform integrates a consortium of three lightweight, fine-tuned, and quantized open-source LLMs -- Gemma, Phi-3.5-mini, and Qwen2 -- selected for their compact architectures and proven efficiency on resource-constrained mobile hardware. An on-device orchestration layer coordinates ensemble inference and consensus-based diagnostic reasoning, producing DSM-5-aligned assessments for conditions. The platform is designed to assist clinicians with differential diagnosis and evidence-linked symptom mapping, as well as to support patient-facing self-screening with appropriate clinical safeguards. Initial evaluation demonstrates that the proposed zero-egress deployment achieves diagnostic accuracy comparable to its server-side predecessor while sustaining real-time inference latency on commodity mobile hardware.

Eranga Bandara, Asanga Gunaratna, Ross Gore +9

Old Dominion University, Norfolk, VA, USA · AI Motion Labs, Melbourne, Australia · Department of Psychiatry and Neurobehavioral Sciences, University of Virginia School of Medicine, Charlottesville, VA, USA +3

On-Device Language Model InferencePrivacy-Preserving Language ModelsMedical DiagnosisClinical Decision SupportMental Health
AskPDF
Apr 20, 2026

Unlocking the Edge deployment and ondevice acceleration of multi-LoRA enabled one-for-all foundational LLM

Deploying large language models (LLMs) on smartphones poses significant engineering challenges due to stringent constraints on memory, latency, and runtime flexibility. In this work, we present a hardware-aware framework for efficient on-device inference of a LLaMA-based multilingual foundation model supporting multiple use cases on Samsung Galaxy S24 and S25 devices with SM8650 and SM8750 Qualcomm chipsets respectively. Our approach integrates application-specific LoRAs as runtime inputs to a single frozen inference graph, enabling dynamic task switching without recompilation or memory overhead. We further introduce a multi-stream decoding mechanism that concurrently generates stylistic variations - such as formal, polite, or jovial responses - within a single forward pass, reducing latency by up to 6x. To accelerate token generation, we apply Dynamic Self-Speculative Decoding (DS2D), a tree-based strategy that predicts future tokens without requiring a draft model, yielding up to 2.3x speedup in decode time. Combined with quantization to INT4 and architecture-level optimizations, our system achieves 4-6x overall improvements in memory and latency while maintaining accuracy across 9 languages and 8 tasks. These results demonstrate practical feasibility of deploying multi-use-case LLMs on edge devices, advancing the commercial viability of Generative AI in mobile platforms.

Sravanth Kodavanti, Sowmya Vajrala, Srinivas Miriyala +13

Samsung Research Institute Bangalore, India · Samsung Electronics, Suwon, South Korea

On-Device Language Model InferenceLLM InferenceSpeculative DecodingLLM Inference Acceleration
AskPDF
Apr 15, 2026

A KL Lens on Quantization: Fast, Forward-Only Sensitivity for Mixed-Precision SSM-Transformer Models

Deploying Large Language Models (LLMs) on edge devices faces severe computational and memory constraints, limiting real-time processing and on-device intelligence. Hybrid architectures combining Structured State Space Models (SSMs) with transformer-based LLMs offer a balance of efficiency and performance. Aggressive quantization can drastically cut model size and speed up inference, but its uneven effects on different components require careful management. In this work, we propose a lightweight, backpropagation-free, surrogate-based sensitivity analysis framework to identify hybrid SSM-Transformer components most susceptible to quantization-induced degradation. Relying solely on forward-pass metrics, our method avoids expensive gradient computations and retraining, making it suitable for situations where access to in-domain data is limited due to proprietary restrictions or privacy constraints. We also provide a formal analysis showing that the Kullback-Leibler (KL) divergence metric better captures quantization sensitivity for Language modeling tasks than widely adopted alternatives such as mean squared error (MSE) and signal-to-quantization-noise ratio (SQNR). Through extensive experiments on SSM and hybrid architectures, our ablation studies confirm that KL-based rankings align with observed performance drops and outperform alternative metrics. This framework enables the practical deployment of advanced hybrid models on resource-constrained edge devices with minimal accuracy loss. We further validate our approach with real-world on-device profiling on Intel Lunar Lake hardware, demonstrating that KL-guided mixed-precision achieves near-FP16 perplexity with model sizes and throughput competitive with Uniform INT4 on both CPU and GPU execution modes. Code is available at https://github.com/jasonkongie/kl-ssm-quant.

Jason Kong, Nilesh Prasad Pandey, Flavio Ponzina +1

Quantization Sensitivity AnalysisLLM QuantizationNeural Network QuantizationMixed-Precision QuantizationOn-Device Language Model Inference
AskPDF
Apr 13, 2026

EdgeCIM: A Hardware-Software Co-Design for CIM-Based Acceleration of Small Language Models

The growing demand for deploying Small Language Models (SLMs) on edge devices, including laptops, smartphones, automotive and robotic computing platforms, and high-performance embedded systems, has exposed fundamental inefficiencies in existing accelerators. While GPUs handle prefill workloads efficiently, the autoregressive decoding phase is dominated by GEMV operations that are inherently memory-bound, resulting in poor utilization and prohibitive energy costs at the edge. In this work, we present EdgeCIM, a hardware-software co-design framework for accelerating autoregressive decoding of SLMs on edge devices in latency-critical applications. At its core is a CIM macro, implemented in 65 nm, coupled with a tile-based mapping strategy that balances pipeline stages, maximizing parallelism while alleviating DRAM bandwidth bottlenecks. Our simulator enables design space exploration of SLMs up to 4B parameters, identifying hardware configurations that optimize latency and energy according to a configurable objective function. On LLaMA3.2 3B (INT4), EdgeCIM achieves 1.76x and 3.21x higher throughput than NVIDIA Jetson AGX Orin and Orin Nano Super operating in MAXN mode, respectively, and 7.07x higher throughput than Qualcomm's Snapdragon-X Elite. Extensive benchmarks on TinyLLaMA 1.1B, LLaMA3.2 (1B, 3B), Phi 3.5 mini 3.8B, Qwen2.5 (0.5B, 1.5B, 3B), SmolLM2 1.7B, SmolLM3 3B, and Qwen3 (0.6B, 1.7B, 4B) reveal that our accelerator, at INT4 precision, achieves on average 336.4 tokens/s and 173 tokens/J. These results establish EdgeCIM as a compelling solution towards real-time, energy-efficient, latency-critical SLM inference at the edge.

Jinane Bazzi, Mariam Rakka, Fadi Kurdahi +2

King Abdullah University of Science and Technology, Thuwal, Saudi Arabia. · the University of California Irvine, Irvine, CA, USA. · Rain Neuromorphics Inc., San Francisco, CA, USA

On-Device Language Model InferenceEfficient Language Model InferenceCompute-in-MemoryHardware-Software Co-DesignSmall Language Models
AskPDF
Apr 8, 2026

SHIELD: A Segmented Hierarchical Memory Architecture for Energy-Efficient LLM Inference on Edge NPUs

Large Language Model (LLM) inference on edge Neural Processing Units (NPUs) is fundamentally constrained by limited on-chip memory capacity. Although high-density embedded DRAM (eDRAM) is attractive for storing activation workspaces, its periodic refresh consumes substantial energy. Prior work has primarily focused on reducing off-chip traffic or optimizing refresh for persistent Key-Value (KV) caches, while transient and error-resilient Query and Attention Output (QO) activations are largely overlooked. We propose SHIELD, a lifecycle-aware segmented eDRAM architecture that jointly exploits temporal residency and bit-level sensitivity in bfloat16 (BF16) activations. SHIELD isolates the sign and exponent fields from the mantissa, disables refresh for transient QO mantissas, and applies relaxed refresh to persistent KV mantissas. Across multiple LLMs and inference scenarios, SHIELD reduces eDRAM refresh energy by 35% relative to a standard-refresh baseline while preserving accuracy on WikiText-2, PIQA, and ARC-Easy.

Jintao Zhang, Xuanyao Fong

Department of Electrical and Computer Engineering, National University of Singapore, Singapore

On-Device Language Model InferenceEfficient Language Model InferenceMemory-Efficient InferenceMemory-Efficient Optimization
AskPDF
Apr 6, 2026

Don't Waste Bits! Adaptive KV-Cache Quantization for Lightweight On-Device LLMs

Large Language Models (LLMs) have achieved remarkable progress across reasoning, generation, and decision-making tasks, yet deploying them on mobile, embedded, and edge devices remains particularly challenging. On-device LLM inference is heavily constrained by the memory and bandwidth overhead of the key-value (KV) cache, which grows linearly with context length and often dominates decoding cost. Existing KV-cache quantization schemes typically rely on fixed precision or hand-crafted heuristics, thereby wasting bits on low-impact tokens while over-compressing informative ones, leading to avoidable accuracy degradation. Inspired by Huffman coding's principle of variable-length allocation, we propose adaptive KV-cache quantization, a learned policy that assigns bit-width proportional to token importance, minimizing expected memory and latency without sacrificing competitive accuracy. Our framework extracts lightweight token-level features, including token frequency, quality score, attention variance, and entropy-based uncertainty, and feeds them into a compact data-driven controller that dynamically selects KV precision from {2-bit, 4-bit, 8-bit, FP16} during decoding. This adaptive precision policy reduces KV memory footprint and latency while improving accuracy compared to static KV quantization and rule-based baselines, and maintaining competitive accuracy close to FP16 inference across standard LLM benchmarks. Extensive experiments across multiple commonsense reasoning benchmarks using SmolLM-135M, SmolLM-360M, and SmolLM-1.7B demonstrate that our controller consistently improves the accuracy-latency trade-off. For instance, with SmolLM-360M on HellaSwag, our method reduces decoding latency (ms/token) by 17.75% relative to static KV quantization, improves accuracy by 7.60 points, and remains within only 0.30 points of FP16 inference.

Sayed Pedram Haeri Boroujeni, Niloufar Mehrabi, Patrick Woods +2

Clemson University, USA

KV-Cache QuantizationOn-Device Language Model InferenceLLM Inference AccelerationMixed-Precision Quantization
AskPDF
Apr 6, 2026

MUXQ: Mixed-to-Uniform Precision MatriX Quantization via Low-Rank Outlier Decomposition

Large language models (LLMs) have achieved outstanding performance across a wide range of natural language processing tasks, but their enormous parameter counts impose ubstantial memory and computational overheads. This challenge is particularly critical in NPU-based on-device environments, where FP16/FP32 computation is inefficient and integer (INT) quantization is therefore essential. However, existing methods, including ZeroQuant, LLM.int8(), and SmoothQuant, do not fully address input-activation outliers and the associated hardware inefficiencies. To overcome these limitations, we propose MUXQ (Mixed-to-Uniform Quantization). MUXQ detects outlier channels in input activations and introduces a small auxiliary matrix that redistributes outlier magnitudes across channels, thereby alleviating the outlier problem. This enables even activation outliers to be quantized at low-precision INT levels while preserving a hardware-friendly computation structure. Experiments on GPT-2 models at three scales (0.1B, 0.3B, and 0.7B parameters) using the WikiText-2 dataset show that MUXQ consistently achieves lower perplexity than naive quantization. In particular, under per-tensor quantization, MUXQ quantizes both activations and weights to INT8 while maintaining accuracy close to that of FP16. With only modest computational overhead, MUXQ enables stable low-precision inference and can be readily combined with other quantization techniques. These results suggest that MUXQ provides a promising direction for efficient and accurate LLM inference on edge devices.

Seoungsub Lee, In Seo Kim, Seon Wook Kim

School of Electrical Engineering, Korea University, Seoul, Korea

LLM QuantizationNeural Network QuantizationActivation QuantizationOn-Device Language Model Inference
AskPDF
Mar 27, 2026

Sustainability Is Not Linear: Quantifying Performance, Energy, and Privacy Trade-offs in On-Device Intelligence

The migration of Large Language Models (LLMs) from cloud clusters to edge devices promises enhanced privacy and offline accessibility, but this transition encounters a harsh reality: the physical constraints of mobile batteries, thermal limits, and, most importantly, memory constraints. To navigate this landscape, we constructed a replicable and reproducible experimental pipeline to profile the complex interplay between energy consumption, latency, and quality of LLMs on mobile devices. We harness this pipeline to conduct an empirical case study on a flagship Android device, capturing granular metrics across eight LLMs ranging from 0.5B to 9B parameters without requiring root access, ensuring our findings reflect realistic user conditions. The findings highlight the trade-offs between generation quality, performance, power and resource consumption, revealing which LLMs offer the best balance across metrics and under different conditions. Besides, we uncovered a counter-intuitive quantization energy paradox: while modern importance-aware quantization successfully reduces memory footprints to fit larger models into RAM, we found it yields negligible energy savings compared to standard mixed-precision methods. This proves that for battery life, the architecture of the model, not its quantization scheme, is the decisive factor. We further identified that Mixture-of-Experts (MoE) architectures defy the standard size-energy trend, offering the storage capacity of a 7B model while maintaining the lower energy profile of a 1B to 2B model. Finally, an analysis of these multi-objective trade-offs reveals a pragmatic sweet spot of mid-sized models, such as Qwen2.5-3B, that effectively balance response quality with sustainable energy consumption.

Eziyo Ehsani, Luca Giamattei, Ivano Malavolta +1

∗University of Naples Federico II, Naples, Italy · †Vrije Universiteit Amsterdam, Amsterdam, The Netherlands

On-Device Language Model InferenceEnergy-Efficient MLLLM EvaluationMixed-Precision Quantization
AskPDF
Mar 24, 2026

PLACID: Privacy-preserving Large language models for Acronym Clinical Inference and Disambiguation

Large Language Models (LLMs) offer transformative solutions across many domains, but healthcare integration is hindered by strict data privacy constraints. Clinical narratives are dense with ambiguous acronyms, misinterpretation these abbreviations can precipitate severe outcomes like life-threatening medication errors. While cloud-dependent LLMs excel at Acronym Disambiguation, transmitting Protected Health Information to external servers violates privacy frameworks. To bridge this gap, this study pioneers the evaluation of small-parameter models deployed entirely on-device to ensure privacy preservation. We introduce a privacy-preserving cascaded pipeline leveraging general-purpose local models to detect clinical acronyms, routing them to domain-specific biomedical models for context-relevant expansions. Results reveal that while general instruction-following models achieve high detection accuracy (~0.988), their expansion capabilities plummet (~0.655). Our cascaded approach utilizes domain-specific medical models to increase expansion accuracy to (~0.81). This novel work demonstrates that privacy-preserving, on-device (2B-10B) models deliver high-fidelity clinical acronym disambiguation support.

Manjushree B. Aithal, Ph. D., Alexander Kotz +2

Department of Biomedical Informatics, University of Colorado Anschutz, Aurora, CO

Clinical NLPPrivacy-Preserving Language ModelsMedical LLMsOn-Device Language Model Inference
AskPDF