Daily Papers

We introduce AIBrix, a cloud-native, open-source framework designed to optimize and simplify large-scale LLM deployment in cloud environments. Unlike traditional cloud-native stacks, AIBrix follows a co-design philosophy, ensuring every layer of the infrastructure is purpose-built for seamless integration with inference engines like vLLM. AIBrix introduces several key innovations to reduce inference costs and enhance performance including high-density LoRA management for dynamic adapter scheduling, LLM-specific autoscalers, and prefix-aware, load-aware routing. To further improve efficiency, AIBrix incorporates a distributed KV cache, boosting token reuse across nodes, leading to a 50% increase in throughput and a 70% reduction in inference latency. AIBrix also supports unified AI runtime which streamlines model management while maintaining vendor-agnostic engine compatibility. For large-scale multi-node inference, AIBrix employs hybrid orchestration -- leveraging Kubernetes for coarse-grained scheduling and Ray for fine-grained execution -- to balance efficiency and flexibility. Additionally, an SLO-driven GPU optimizer dynamically adjusts resource allocations, optimizing heterogeneous serving to maximize cost efficiency while maintaining service guarantees. Finally, AIBrix enhances system reliability with AI accelerator diagnostic tools, enabling automated failure detection and mock-up testing to improve fault resilience. AIBrix is available at https://github.com/vllm-project/aibrix.

Optimizing GPU kernels manually is a challenging and time-consuming task. With the rapid development of LLMs, automated GPU kernel optimization is gradually becoming a tangible reality. However, current LLM-driven automated optimization methods narrowly focus on machine learning applications, such as PyTorch operator optimization, while overlooking broader domains like sparse matrix operations in scientific computing. Extending to these broader applications brings new challenges for the benchmark and algorithm. Therefore, developing a general-purpose automated kernel optimization method becomes our primary focus. In this paper, we address the absence of systematic evaluation for multi-scenario settings by introducing MSKernelBench, which spans multiple scenarios, including fundamental algebraic operations, common LLM kernels, sparse matrix operators, and scientific computing routines, each supporting both FP32 and BF16 precision. Building on this benchmark, we introduce CUDAMaster, a multi-agent, hardware-aware system for kernel optimization that leverages profiling information and automatically constructs the full compilation and execution toolchain. Experimental results demonstrate that CUDAMaster achieves significant speedups across most operators, outperforming Astra by about 35%. In several cases, its performance matches or surpasses that of highly optimized, closed-source libraries such as cuBLAS. A demo showcasing the original and optimized code for each operator is available at https://hanyx2021.github.io/MSKernelBenchDemo/.

Large language models (LLMs) are increasingly integrated into many online services. However, a major challenge in deploying LLMs is their high cost, due primarily to the use of expensive GPU instances. To address this problem, we find that the significant heterogeneity of GPU types presents an opportunity to increase GPU cost efficiency and reduce deployment costs. The broad and growing market of GPUs creates a diverse option space with varying costs and hardware specifications. Within this space, we show that there is not a linear relationship between GPU cost and performance, and identify three key LLM service characteristics that significantly affect which GPU type is the most cost effective: model request size, request rate, and latency service-level objective (SLO). We then present M\'elange, a framework for navigating the diversity of GPUs and LLM service specifications to derive the most cost-efficient set of GPUs for a given LLM service. We frame the task of GPU selection as a cost-aware bin-packing problem, where GPUs are bins with a capacity and cost, and items are request slices defined by a request size and rate. Upon solution, M\'elange derives the minimal-cost GPU allocation that adheres to a configurable latency SLO. Our evaluations across both real-world and synthetic datasets demonstrate that M\'elange can reduce deployment costs by up to 77% as compared to utilizing only a single GPU type, highlighting the importance of making heterogeneity-aware GPU provisioning decisions for LLM serving. Our source code is publicly available at https://github.com/tyler-griggs/melange-release.

Fine-tuning large pre-trained LLMs generally demands extensive GPU memory. Traditional first-order optimizers like SGD encounter substantial difficulties due to increased memory requirements from storing activations and gradients during both the forward and backward phases as the model size expands. Alternatively, zeroth-order (ZO) techniques can compute gradients using just forward operations, eliminating the need to store activations. Furthermore, by leveraging CPU capabilities, it's feasible to enhance both the memory and processing power available to a single GPU. We propose a novel framework, ZO2 (Zeroth-Order Offloading), for efficient zeroth-order fine-tuning of LLMs with only limited GPU memory. Our framework dynamically shifts model parameters between the CPU and GPU as required, optimizing computation flow and maximizing GPU usage by minimizing downtime. This integration of parameter adjustments with ZO's double forward operations reduces unnecessary data movement, enhancing the fine-tuning efficacy. Additionally, our framework supports an innovative low-bit precision approach in AMP mode to streamline data exchanges between the CPU and GPU. Employing this approach allows us to fine-tune extraordinarily large models, such as the OPT-175B with more than 175 billion parameters, on a mere 18GB GPU--achievements beyond the reach of traditional methods. Moreover, our framework achieves these results with almost no additional time overhead and absolutely no accuracy loss compared to standard zeroth-order methods. ZO2's code has been open-sourced in https://github.com/liangyuwang/zo2.

The exponential growth in demand for GPU computing resources, driven by the rapid advancement of Large Language Models, has created an urgent need for automated CUDA optimization strategies. While recent advances in LLMs show promise for code generation, current SOTA models (e.g. R1, o1) achieve low success rates in improving CUDA speed. In this paper, we introduce CUDA-L1, an automated reinforcement learning framework for CUDA optimization. CUDA-L1 achieves performance improvements on the CUDA optimization task: trained on NVIDIA A100, it delivers an average speedup of x17.7 across all 250 CUDA kernels of KernelBench, with peak speedups reaching x449. Furthermore, the model also demonstrates excellent portability across GPU architectures, achieving average speedups of x17.8 on H100, x19.0 on RTX 3090, x16.5 on L40, x14.7 on H800, and x13.9 on H20 despite being optimized specifically for A100. Beyond these benchmark results, CUDA-L1 demonstrates several remarkable properties: 1) Discovers a variety of CUDA optimization techniques and learns to combine them strategically to achieve optimal performance; 2) Uncovers fundamental principles of CUDA optimization; 3) Identifies non-obvious performance bottlenecks and rejects seemingly beneficial optimizations that harm performance. The capabilities of CUDA-L1 demonstrate that reinforcement learning can transform an initially poor-performing LLM into an effective CUDA optimizer through speedup-based reward signals alone, without human expertise or domain knowledge. More importantly, the trained RL model extend the acquired reasoning abilities to new kernels. This paradigm opens possibilities for automated optimization of CUDA operations, and holds promise to substantially promote GPU efficiency and alleviate the rising pressure on GPU computing resources.

Monolithic serving with chunked prefill improves GPU utilization by batching prefill and decode together, but suffers from fine-grained phase interference. Engine-level prefill-decode (PD) disaggregation avoids interference but incurs higher hardware and coordination overhead. Prior intra-GPU disaggregation approaches multiplex prefill and decode within a single GPU, using SLO-based tuning guided by heuristics from offline profiling or reactive feedback loops. However, these methods respond reactively to performance issues rather than anticipating them, limiting adaptability under dynamic workloads. We ask: can we achieve proactive intra-GPU disaggregation that adapts effectively to dynamic workloads? The key challenge lies in managing the conflicting resource demands of prefill and decode under varying conditions. We first show that GPU resources exhibit diminishing returns -- beyond a saturation point, more allocation yields minimal latency benefit. Second, we observe that memory bandwidth contention becomes a critical bottleneck. These insights motivate a design that dynamically partitions GPU resources across prefill and decode phases, while jointly considering compute capacity, memory footprint, and bandwidth contention. Evaluated on diverse LLMs and workloads, our system Nexus achieves up to 2.2x higher throughput, 20x lower TTFT, and 2.5x lower TBT than vLLM; outperforms SGLang by up to 2x; and matches or exceeds disaggregated vLLM.

Generating high-performance CUDA kernels remains challenging due to the need to navigate a combinatorial space of low-level transformations under noisy and expensive hardware feedback. Although large language models can synthesize functionally correct CUDA code, achieving competitive performance requires systematic exploration and verification of optimization choices. We present OptiML, an end-to-end framework that maps either natural-language intent or input CUDA code to performance-optimized CUDA kernels by formulating kernel optimization as search under verification. OptiML consists of two decoupled stages. When the input is natural language, a Mixture-of-Thoughts generator (OptiML-G) acts as a proposal policy over kernel implementation strategies, producing an initial executable program. A search-based optimizer (OptiML-X) then refines either synthesized or user-provided kernels using Monte Carlo Tree Search over LLM-driven edits, guided by a hardware-aware reward derived from profiler feedback. Each candidate transformation is compiled, verified, and profiled with Nsight Compute, and evaluated by a composite objective that combines runtime with hardware bottleneck proxies and guardrails against regressions. We evaluate OptiML in both synthesis-and-optimize and optimization-only settings on a diverse suite of CUDA kernels. Results show that OptiML consistently discovers verified performance improvements over strong LLM baselines and produces interpretable optimization trajectories grounded in profiler evidence.

Large Language Model (LLM) serving faces a fundamental tension between stringent latency Service Level Objectives (SLOs) and limited GPU memory capacity. When high request rates exhaust the KV cache budget, existing LLM inference systems often suffer severe head-of-line (HOL) blocking. While prior work explored PCIe-based offloading, these approaches cannot sustain responsiveness under high request rates, often failing to meet tight Time-To-First-Token (TTFT) and Time-Between-Tokens (TBT) SLOs. We present SuperInfer, a high-performance LLM inference system designed for emerging Superchips (e.g., NVIDIA GH200) with tightly coupled GPU-CPU architecture via NVLink-C2C. SuperInfer introduces RotaSched, the first proactive, SLO-aware rotary scheduler that rotates requests to maintain responsiveness on Superchips, and DuplexKV, an optimized rotation engine that enables full-duplex transfer over NVLink-C2C. Evaluations on GH200 using various models and datasets show that SuperInfer improves TTFT SLO attainment rates by up to 74.7% while maintaining comparable TBT and throughput compared to state-of-the-art systems, demonstrating that SLO-aware scheduling and memory co-design unlocks the full potential of Superchips for responsive LLM serving.

GPU kernel optimization has long been a central challenge at the intersection of high-performance computing and machine learning. Efficient kernels are crucial for accelerating large language model (LLM) training and serving, yet attaining high performance typically requires extensive manual tuning. Compiler-based systems reduce some of this burden, but still demand substantial manual design and engineering effort. Recently, researchers have explored using LLMs for GPU kernel generation, though prior work has largely focused on translating high-level PyTorch modules into CUDA code. In this work, we introduce Astra, the first LLM-based multi-agent system for GPU kernel optimization. Unlike previous approaches, Astra starts from existing CUDA implementations extracted from SGLang, a widely deployed framework for serving LLMs, rather than treating PyTorch modules as the specification. Within Astra, specialized LLM agents collaborate through iterative code generation, testing, profiling, and planning to produce kernels that are both correct and high-performance. On kernels from SGLang, Astra achieves an average speedup of 1.32x using zero-shot prompting with OpenAI o4-mini. A detailed case study further demonstrates that LLMs can autonomously apply loop transformations, optimize memory access patterns, exploit CUDA intrinsics, and leverage fast math operations to yield substantial performance gains. Our work highlights multi-agent LLM systems as a promising new paradigm for GPU kernel optimization. Our code is publicly available at https://github.com/Anjiang-Wei/Astra.

Large Language Models (LLMs) have demonstrated strong capabilities in general-purpose code generation. However, generating the code which is deeply hardware-specific, architecture-aware, and performance-critical, especially for massively parallel GPUs, remains a complex challenge. In this work, we explore the use of LLMs for the automated generation and optimization of CUDA programs, with the goal of producing high-performance GPU kernels that fully exploit the underlying hardware. To address this challenge, we propose a novel framework called Feature Search and Reinforcement (FSR). FSR jointly optimizes compilation and functional correctness, as well as the runtime performance, which are validated through extensive and diverse test cases, and measured by actual kernel execution latency on the target GPU, respectively. This approach enables LLMs not only to generate syntactically and semantically correct CUDA code but also to iteratively refine it for efficiency, tailored to the characteristics of the GPU architecture. We evaluate FSR on representative CUDA kernels, covering AI workloads and computational intensive algorithms. Our results show that LLMs augmented with FSR consistently guarantee correctness rates. Meanwhile, the automatically generated kernels can outperform general human-written code by a factor of up to 179times in execution speeds. These findings highlight the potential of combining LLMs with performance reinforcement to automate GPU programming for hardware-specific, architecture-sensitive, and performance-critical applications.

Two widely adopted techniques for LLM inference serving systems today are hybrid batching and disaggregated serving. A hybrid batch combines prefill and decode tokens of different requests in the same batch to improve resource utilization and throughput at the cost of increased latency per token. In contrast, disaggregated serving decouples compute-bound prefill and bandwidth-bound decode phases to optimize for service level objectives (SLOs) at the cost of resource under-utilization and KV-cache transfer overheads. To address the limitations of these techniques, we propose RAPID-Serve: a technique to concurrently execute prefill and decode on the same GPU(s) to meet latency SLOs while maintaining high throughput and efficient resource utilization. Furthermore, we propose Adaptive Resource Management for runtime compute resource allocation, optionally leveraging CU masking (a fine-grained Compute Unit partitioning feature on AMD InstinctTM GPUs). RAPID-Serve provides up to 4.1x (average 1.7x) unconstrained throughput improvement and 32x and higher (average 4.9x) throughput improvement under SLO constraints, showing it as an effective strategy compared to the state-of-the-art approaches, particularly in resource-constrained environments.

The recent shift in Generative AI (GenAI) applications from cloud-only environments to end-user devices introduces new challenges in resource management, system efficiency, and user experience. This paper presents ConsumerBench, a comprehensive benchmarking framework designed to evaluate the system efficiency and response time of GenAI models running on end-user devices. Unlike existing benchmarks that assume exclusive model access on dedicated GPUs, ConsumerBench simulates realistic multi-application scenarios executing concurrently on constrained hardware. Furthermore, ConsumerBench supports customizable workflows that simulate complex tasks requiring coordination among multiple applications. ConsumerBench captures both application-level metrics, including latency and Service Level Objective (SLO) attainment, and system-level metrics like CPU/GPU utilization and memory bandwidth. Through extensive experiments, ConsumerBench reveals inefficiencies in resource sharing, unfair scheduling under greedy allocation, and performance pitfalls of static model server configurations. The paper also provides practical insights for model developers and system designers, highlighting the benefits of custom kernels tailored to consumer-grade GPU architectures and the value of implementing SLO-aware scheduling strategies.

Enterprise LLM deployment faces a critical scalability challenge: organizations must optimize models systematically to scale AI initiatives within constrained compute budgets, yet the specialized expertise required for manual optimization remains a niche and scarce skillset. This challenge is particularly evident in managing GPU utilization across heterogeneous infrastructure while enabling teams with diverse workloads and limited LLM optimization experience to deploy models efficiently. We present OptiKIT, a distributed LLM optimization framework that democratizes model compression and tuning by automating complex optimization workflows for non-expert teams. OptiKIT provides dynamic resource allocation, staged pipeline execution with automatic cleanup, and seamless enterprise integration. In production, it delivers more than 2x GPU throughput improvement while empowering application teams to achieve consistent performance improvements without deep LLM optimization expertise. We share both the platform design and key engineering insights into resource allocation algorithms, pipeline orchestration, and integration patterns that enable large-scale, production-grade democratization of model optimization. Finally, we open-source the system to enable external contributions and broader reproducibility.

Optimizing GPU kernels is critical for efficient modern machine learning systems yet remains challenging due to the complex interplay of design factors and rapid hardware evolution. Existing automated approaches typically treat Large Language Models (LLMs) merely as stochastic code generators within heuristic-guided evolutionary loops. These methods often struggle with complex kernels requiring coordinated, multi-step structural transformations, as they lack explicit planning capabilities and frequently discard promising strategies due to inefficient or incorrect intermediate implementations. To address this, we propose Search via Co-Evolving World Model and build K-Search based on this method. By replacing static search heuristics with a co-evolving world model, our framework leverages LLMs' prior domain knowledge to guide the search, actively exploring the optimization space. This approach explicitly decouples high-level algorithmic planning from low-level program instantiation, enabling the system to navigate non-monotonic optimization paths while remaining resilient to temporary implementation defects. We evaluate K-Search on diverse, complex kernels from FlashInfer, including GQA, MLA, and MoE kernels. Our results show that K-Search significantly outperforms state-of-the-art evolutionary search methods, achieving an average 2.10x improvement and up to a 14.3x gain on complex MoE kernels. On the GPUMode TriMul task, K-Search achieves state-of-the-art performance on H100, reaching 1030us and surpassing both prior evolution and human-designed solutions.

GPU kernel optimization is fundamental to modern deep learning but remains a highly specialized task requiring deep hardware expertise. Despite strong performance in general programming, large language models (LLMs) remain uncompetitive with compiler-based systems such as torch.compile for CUDA kernel generation. Existing CUDA code generation approaches either rely on training-free refinement or fine-tune models within fixed multi-turn execution-feedback loops, but both paradigms fail to fundamentally improve the model's intrinsic CUDA optimization ability, resulting in limited performance gains. We present CUDA Agent, a large-scale agentic reinforcement learning system that develops CUDA kernel expertise through three components: a scalable data synthesis pipeline, a skill-augmented CUDA development environment with automated verification and profiling to provide reliable reward signals, and reinforcement learning algorithmic techniques enabling stable training. CUDA Agent achieves state-of-the-art results on KernelBench, delivering 100\%, 100\%, and 92\% faster rate over torch.compile on KernelBench Level-1, Level-2, and Level-3 splits, outperforming the strongest proprietary models such as Claude Opus 4.5 and Gemini 3 Pro by about 40\% on the hardest Level-3 setting.

Optimizing GPU kernels presents a significantly greater challenge for large language models (LLMs) than standard code generation tasks, as it requires understanding hardware architecture, parallel optimization strategies, and performance profiling outputs. Most existing LLM-based approaches to kernel generation rely on simple prompting and feedback loops, incorporating hardware awareness only indirectly through profiling feedback. We introduce KernelFoundry, an evolutionary framework that efficiently explores the GPU kernel design space through three key mechanisms: (1) MAP-Elites quality-diversity search with kernel-specific behavioral dimensions to sustain exploration across diverse optimization strategies; (2) meta-prompt evolution, which co-evolves prompts with kernels to uncover task-specific optimization strategies, and (3) template-based parameter optimization to tune kernels to inputs and hardware. We evaluate this framework on KernelBench, robust-kbench, and custom tasks, generating SYCL kernels as a cross-platform GPU programming model and CUDA kernels for comparison to prior work. Our approach consistently outperforms the baseline methods, achieving an average speedup of 2.3x on KernelBench for SYCL. Moreover, KernelFoundry is implemented as a distributed framework with remote access to diverse hardware, enabling rapid benchmarking and featuring a flexible user input layer that supports kernel generation for a wide range of real-world use cases beyond benchmarking.

As agentic AI systems become increasingly capable of generating and optimizing GPU kernels, progress is constrained by benchmarks that reward speedup over software baselines rather than proximity to hardware-efficient execution. We present SOL-ExecBench, a benchmark of 235 CUDA kernel optimization problems extracted from 124 production and emerging AI models spanning language, diffusion, vision, audio, video, and hybrid architectures, targeting NVIDIA Blackwell GPUs. The benchmark covers forward and backward workloads across BF16, FP8, and NVFP4, including kernels whose best performance is expected to rely on Blackwell-specific capabilities. Unlike prior benchmarks that evaluate kernels primarily relative to software implementations, SOL-ExecBench measures performance against analytically derived Speed-of-Light (SOL) bounds computed by SOLAR, our pipeline for deriving hardware-grounded SOL bounds, yielding a fixed target for hardware-efficient optimization. We report a SOL Score that quantifies how much of the gap between a release-defined scoring baseline and the hardware SOL bound a candidate kernel closes. To support robust evaluation of agentic optimizers, we additionally provide a sandboxed harness with GPU clock locking, L2 cache clearing, isolated subprocess execution, and static analysis based checks against common reward-hacking strategies. SOL-ExecBench reframes GPU kernel benchmarking from beating a mutable software baseline to closing the remaining gap to hardware Speed-of-Light.

In this paper, we propose CUDA-L2, a system that combines large language models (LLMs) and reinforcement learning (RL) to automatically optimize Half-precision General Matrix Multiply (HGEMM) CUDA kernels. Using CUDA execution speed as the RL reward, CUDA-L2 automatically optimizes HGEMM kernels across 1,000 configurations. CUDA-L2 systematically outperforms major matmul baselines to date, from the widely-used {\it torch.matmul} to state-of-the-art Nvidia's closed-source libraries, i.e., {\it cuBLAS}, {\it cuBLASLt}. In offline mode, where kernels are executed consecutively without time intervals, CUDA-L2 yields +22.0\% over {\it torch.matmul} on average; +19.2\% over {\it cuBLAS} using the optimal layout configuration (normal-normal NN and transposed-normal TN); +16.8\% over {\it cuBLASLt-heuristic}, which queries {\it cuBLASLt} library and selects the algorithm based on the heuristic's suggestion; and +11.4\% over the most competitive {\it cuBLASLt-AutoTuning} model, which selects the fastest algorithm from up to 100 candidates from {\it cuBLASLt}'s suggestions. In server mode, where kernels are executed at random intervals simulating real-time inference, the speedups further increase to +28.7\%, +26.0\%, +22.4\%, and +15.9\% for {\it torch.matmul}, {\it cuBLAS}, {\it cuBLASLt-heuristic}, and {\it cuBLASLt-AutoTuning} respectively. CUDA-L2 shows that even the most performance-critical, heavily-optimized kernels like HGEMM can be improved through LLM-guided RL automation by systematically exploring configuration spaces at scales impractical for humans. Project and code can be found at github.com/deepreinforce-ai/CUDA-L2

The efficiency of GPU kernels is central to the progress of modern AI, yet optimizing them remains a difficult and labor-intensive task due to complex interactions between memory hierarchies, thread scheduling, and hardware-specific characteristics. While recent advances in large language models (LLMs) provide new opportunities for automated code generation, existing approaches largely treat LLMs as single-shot generators or naive refinement tools, limiting their effectiveness in navigating the irregular kernel optimization landscape. We introduce an LLM agentic framework for GPU kernel optimization that systematically explores the design space through multi-agent collaboration, grounded instruction, dynamic context management, and strategic search. This framework mimics the workflow of expert engineers, enabling LLMs to reason about hardware trade-offs, incorporate profiling feedback, and refine kernels iteratively. We evaluate our approach on KernelBench, a benchmark for LLM-based kernel optimization, and demonstrate substantial improvements over baseline agents: our system produces correct solutions where baselines often fail, and achieves kernels with up to 16x faster runtime performance. These results highlight the potential of agentic LLM frameworks to advance fully automated, scalable GPU kernel optimization.

Large language models (LLMs) are remarked by their substantial computational requirements. To mitigate the cost, researchers develop specialized CUDA kernels, which often fuse several tensor operations to maximize the utilization of GPUs as much as possible. However, those specialized kernels may still leave performance on the table as CUDA assembly experts show that manual optimization of GPU SASS schedules can lead to better performance, and trial-and-error is largely employed to manually find the best GPU SASS schedules. In this work, we employ an automatic approach to optimize GPU SASS schedules, which thus can be integrated into existing compiler frameworks. The key to automatic optimization is training an RL agent to mimic how human experts perform manual scheduling. To this end, we formulate an assembly game, where RL agents can play to find the best GPU SASS schedules. The assembly game starts from a -O3 optimized SASS schedule, and the RL agents can iteratively apply actions to mutate the current schedules. Positive rewards are generated if the mutated schedules get higher throughput by executing on GPUs. Experiments show that CuAsmRL can further improve the performance of existing specialized CUDA kernels transparently by up to 26%, and on average 9%. Moreover, it is used as a tool to reveal potential optimization moves learned automatically.

Modern machine learning (ML) workloads increasingly rely on GPUs, yet achieving high end-to-end performance remains challenging due to dependencies on both GPU kernel efficiency and host-side settings. Although LLM-based methods show promise on automated GPU kernel generation, prior works mainly focus on single-kernel optimization and do not extend to end-to-end programs, hindering practical deployment. To address the challenge, in this work, we propose StitchCUDA, a multi-agent framework for end-to-end GPU program generation, with three specialized agents: a Planner to orchestrate whole system design, a Coder dedicated to implementing it step-by-step, and a Verifier for correctness check and performance profiling using Nsys/NCU. To fundamentally improve the Coder's ability in end-to-end GPU programming, StitchCUDA integrates rubric-based agentic reinforcement learning over two atomic skills, task-to-code generation and feedback-driven code optimization, with combined rubric reward and rule-based reward from real executions. Therefore, the Coder learns how to implement advanced CUDA programming techniques (e.g., custom kernel fusion, cublas epilogue), and we also effectively prevent Coder's reward hacking (e.g., just copy PyTorch code or hardcoding output) during benchmarking. Experiments on KernelBench show that StitchCUDA achieves nearly 100% success rate on end-to-end GPU programming tasks, with 1.72x better speedup over the multi-agent baseline and 2.73x than the RL model baselines.

Developing high-performance GPU kernels is critical for AI and scientific computing, but remains challenging due to its reliance on expert crafting and poor portability. While LLMs offer promise for automation, both general-purpose and finetuned LLMs suffer from two fundamental and conflicting limitations: correctness and efficiency. The key reason is that existing LLM-based approaches directly generate the entire optimized low-level programs, requiring exploration of an extremely vast space encompassing both optimization policies and implementation codes. To address the challenge of exploring an intractable space, we propose Macro Thinking Micro Coding (MTMC), a hierarchical framework inspired by the staged optimization strategy of human experts. It decouples optimization strategy from implementation details, ensuring efficiency through high-level strategy and correctness through low-level implementation. Specifically, Macro Thinking employs reinforcement learning to guide lightweight LLMs in efficiently exploring and learning semantic optimization strategies that maximize hardware utilization. Micro Coding leverages general-purpose LLMs to incrementally implement the stepwise optimization proposals from Macro Thinking, avoiding full-kernel generation errors. Together, they effectively navigate the vast optimization space and intricate implementation details, enabling LLMs for high-performance GPU kernel generation. Comprehensive results on widely adopted benchmarks demonstrate the superior performance of MTMC on GPU kernel generation in both accuracy and running time. On KernelBench, MTMC achieves near 100% and 70% accuracy at Levels 1-2 and 3, over 50% than SOTA general-purpose and domain-finetuned LLMs, with up to 7.3x speedup over LLMs, and 2.2x over expert-optimized PyTorch Eager kernels. On the more challenging TritonBench, MTMC attains up to 59.64% accuracy and 34x speedup.

Recent studies have demonstrated the potential of Large Language Models (LLMs) in generating GPU Kernels. Current benchmarks focus on the translation of high-level languages into CUDA, overlooking the more general and challenging task of text-to-CUDA generation. Furthermore, given the hardware-specific and performance-critical features of GPU programming, accurately assessing the performance of LLM-generated GPU programs is nontrivial. In this work, we introduce CUDABench, a comprehensive benchmark designed to evaluate the text-to-CUDA capabilities of LLMs. First, we construct CUDABench-Set, which covers Breadth-Depth-Difficulty evaluation space in diverse application domains, including artificial intelligence, scientific computing, and data analytics, etc. Furthermore, we propose CUDABench-Score and Generative Verification Pipeline that assess (1) compilation correctness, (2) functional consistency through execution-based verification, and (3) a novel roofline-based metric, Performance-Score. Benchmarking state-of-the-art LLMs reveals insightful findings and challenges of text-to-CUDA, such as a notable mismatch between high compilation success rates and low functional correctness, a lack of domain-specific algorithmic knowledge, and suboptimal utilization of GPU hardware resources. Our benchmark is available at https://github.com/CUDA-Bench/CUDABench.

Recent advances in large language models (LLMs) demonstrate their effectiveness in scaling test-time compute for software engineering tasks. However, these approaches often focus on high-level solutions, with limited attention to optimizing low-level CUDA kernel implementations. Additionally, existing kernel generation benchmarks suffer from exploitable loopholes and insufficient diversity in testing conditions, hindering true generalization assessment. To address these limitations, we introduce robust-kbench, a new benchmark for rigorous evaluation of kernel performance and correctness across varied scenarios. Furthermore, we present a comprehensive agentic framework that automates CUDA kernel discovery, verification, and optimization. This pipeline enables frontier LLMs to translate torch code to CUDA kernels and iteratively improve their runtime within our robust evaluation setting. Our sequential workflow first translates PyTorch code into equivalent CUDA kernels. It then optimizes their runtime using a novel evolutionary meta-generation procedure tailored to the CUDA ecosystem, guided by LLM-based verifiers for correctness and efficient filtering. Evaluated on robust-kbench, our approach produces CUDA kernels outperforming torch implementations for practical applications, including forward and backward passes. It can fuse operations and deploy various runtime optimization strategies. The verifier workflow accurately classifies incorrect kernels, enhancing hardware verification efficiency.

We present Kernel-Smith, a framework for high-performance GPU kernel and operator generation that combines a stable evaluation-driven evolutionary agent with an evolution-oriented post-training recipe. On the agent side, Kernel-Smith maintains a population of executable candidates and iteratively improves them using an archive of top-performing and diverse programs together with structured execution feedback on compilation, correctness, and speedup. To make this search reliable, we build backend-specific evaluation services for Triton on NVIDIA GPUs and Maca on MetaX GPUs. On the training side, we convert long-horizon evolution trajectories into step-centric supervision and reinforcement learning signals by retaining correctness-preserving, high-gain revisions, so that the model is optimized as a strong local improver inside the evolutionary loop rather than as a one-shot generator. Under a unified evolutionary protocol, Kernel-Smith-235B-RL achieves state-of-the-art overall performance on KernelBench with Nvidia Triton backend, attaining the best average speedup ratio and outperforming frontier proprietary models including Gemini-3.0-pro and Claude-4.6-opus. We further validate the framework on the MetaX MACA backend, where our Kernel-Smith-MACA-30B surpasses large-scale counterparts such as DeepSeek-V3.2-think and Qwen3-235B-2507-think, highlighting potential for seamless adaptation across heterogeneous platforms. Beyond benchmark results, the same workflow produces upstream contributions to production systems including SGLang and LMDeploy, demonstrating that LLM-driven kernel optimization can transfer from controlled evaluation to practical deployment.

In the context of Machine Learning as a Service (MLaaS) clouds, the extensive use of Large Language Models (LLMs) often requires efficient management of significant query loads. When providing real-time inference services, several challenges arise. Firstly, increasing the number of GPUs may lead to a decrease in inference speed due to heightened communication overhead, while an inadequate number of GPUs can lead to out-of-memory errors. Secondly, different deployment strategies need to be evaluated to guarantee optimal utilization and minimal inference latency. Lastly, inefficient orchestration of inference queries can easily lead to significant Service Level Objective (SLO) violations. Lastly, inefficient orchestration of inference queries can easily lead to significant Service Level Objective (SLO) violations. To address these challenges, we propose a Unified and Efficient approach for Large Language Model inference serving (UELLM), which consists of three main components: 1) resource profiler, 2) batch scheduler, and 3) LLM deployer. UELLM minimizes resource overhead, reduces inference latency, and lowers SLO violation rates. Compared with state-of-the-art (SOTA) techniques, UELLM reduces the inference latency by 72.3% to 90.3%, enhances GPU utilization by 1.2X to 4.1X, and increases throughput by 1.92X to 4.98X, it can also serve without violating the inference latency SLO.

Shampoo is one of the leading approximate second-order optimizers: a variant of it has won the MLCommons AlgoPerf competition, and it has been shown to produce models with lower activation outliers that are easier to compress. Yet, applying Shampoo currently comes at the cost of significant computational slowdown, due to its expensive internal operations. In this paper, we take a significant step to address this shortcoming by proposing \method (for Distributed Accelerated SHampoo), a faster implementation of Distributed Shampoo based on two main new techniques: First, we show that preconditioner blocks can be stacked into 3D tensors to significantly improve GPU utilization; second, we introduce the Newton-DB iteration and the Chebyshev polynomial approximations as novel and faster approaches for computing the inverse matrix roots required by Shampoo. Along with these algorithmic contributions, we provide a first in-depth analysis of how matrix scaling critically affects Shampoo convergence. On the practical side, our GPU-aware implementation achieves up to 4.83times faster optimizer steps compared to the well-optimized Distributed Shampoo, while Newton-DB attains the lowest validation perplexity per iteration among all tested methods. Our code is available at https://github.com/IST-DASLab/DASH.

The development of large language models (LLMs) has been instrumental in advancing state-of-the-art natural language processing applications. Training LLMs with billions of parameters and trillions of tokens require sophisticated distributed systems that enable composing and comparing several state-of-the-art techniques in order to efficiently scale across thousands of accelerators. However, existing solutions are complex, scattered across multiple libraries/repositories, lack interoperability, and are cumbersome to maintain. Thus, curating and empirically comparing training recipes require non-trivial engineering effort. This paper introduces TorchTitan, an open-source, PyTorch-native distributed training system that unifies state-of-the-art techniques, streamlining integration and reducing overhead. TorchTitan enables 3D parallelism in a modular manner with elastic scaling, providing comprehensive logging, checkpointing, and debugging tools for production-ready training. It also incorporates hardware-software co-designed solutions, leveraging features like Float8 training and SymmetricMemory. As a flexible test bed, TorchTitan facilitates custom recipe curation and comparison, allowing us to develop optimized training recipes for Llama 3.1 and provide guidance on selecting techniques for maximum efficiency based on our experiences. We thoroughly assess TorchTitan on the Llama 3.1 family of LLMs, spanning 8 billion to 405 billion parameters, and showcase its exceptional performance, modular composability, and elastic scalability. By stacking training optimizations, we demonstrate accelerations of 65.08% with 1D parallelism at the 128-GPU scale (Llama 3.1 8B), an additional 12.59% with 2D parallelism at the 256-GPU scale (Llama 3.1 70B), and an additional 30% with 3D parallelism at the 512-GPU scale (Llama 3.1 405B) on NVIDIA H100 GPUs over optimized baselines.

LAPS identifies and disaggregates requests with different prompt lengths in LLM serving to reduce TTFT latency. While recent systems have decoupled the prefill and decode stages to improve throughput, they still rely on unified scheduling policies that fail to adapt to heterogeneous workload characteristics. We observe that prompt-length variations lead to distinct performance bottlenecks, motivating an adaptive scheduling strategy. LAPS disaggregates multi-turn long-prefill requests from short-prefill ones and introduces a length-aware smart batching mechanism for short-prefill workloads. It adopts a dual-queue design that supports temporal disaggregation on a single prefill instance or spatial disaggregation across multiple instances. For short-prefill batches, a batch waiting window and CUDA Graph-based clustering mitigate interference from heterogeneous computation, reducing batching delay and lowering average latency. In real multi-turn workloads, LAPS reduces prefill latency by over 30\% compared to vanilla SGLang under prefill-decode disaggregation, and further decreases SLO violations by 28\% in multi-instance deployments with vanilla data-parallel configuration. Compared to the SGLang router with load balancing, it further lowers SLO violations by 12\% in multi-GPU settings. Under high concurrency and mixed-request scenarios, LAPS improves request throughput by 35\% serving Qwen2.5-32B model for prefill instance, demonstrating its effectiveness in optimizing heterogeneous LLM serving workloads.

Fine-tuning large language models (LLMs) with parameter-efficient techniques such as LoRA and QLoRA has enabled adaptation of foundation models on modest hardware. Yet the efficiency of such training on consumer-grade GPUs, especially under strict 8 GB VRAM limits, remains underexplored. We present a controlled profiling study of LoRA/QLoRA fine-tuning using the Qwen2.5-1.5B-Instruct model on a single NVIDIA RTX 4060. Across three representative configurations, we systematically vary batch size, sequence length, optimizer choice (AdamW vs. PagedAdamW), and precision (fp16 vs. bf16). We report throughput (tokens/s), time per 10k tokens, and VRAM footprint, alongside energy estimates derived from GPU board power limits. Our results show that paged optimizers improve throughput by up to 25% (628 tok/s vs. 500 tok/s baseline), while bf16 degrades efficiency relative to fp16. Despite 8 GB constraints, sequence lengths up to 2048 tokens were feasible using parameter-efficient strategies. To our knowledge, this is the first systematic case study of LLM fine- tuning efficiency on consumer GPUs, providing reproducible benchmarks and practical guidelines for resource-constrained researchers and practitioners.

Learned optimization has emerged as a promising alternative to hand-crafted optimizers, with the potential to discover stronger learned update rules that enable faster, hyperparameter-free training of neural networks. A critical element for practically useful learned optimizers, that can be used off-the-shelf after meta-training, is strong meta-generalization: the ability to apply the optimizers to new tasks. Recent state-of-the-art work in learned optimizers, VeLO (Metz et al., 2022), requires a large number of highly diverse meta-training tasks along with massive computational resources, 4000 TPU months, to achieve meta-generalization. This makes further improvements to such learned optimizers impractical. In this work, we identify several key elements in learned optimizer architectures and meta-training procedures that can lead to strong meta-generalization. We also propose evaluation metrics to reliably assess quantitative performance of an optimizer at scale on a set of evaluation tasks. Our proposed approach, Celo, makes a significant leap in improving the meta-generalization performance of learned optimizers and also outperforms tuned state-of-the-art optimizers on a diverse set of out-of-distribution tasks, despite being meta-trained for just 24 GPU hours.

Optimizing CUDA code across multiple generations of GPU architectures is challenging, as achieving peak performance requires an extensive exploration of an increasingly complex, hardware-specific optimization space. Traditional compilers are constrained by fixed heuristics, whereas finetuning Large Language Models (LLMs) can be expensive. However, agentic workflows for CUDA code optimization have limited ability to aggregate knowledge from prior exploration, leading to biased sampling and suboptimal solutions. We propose KernelBlaster, a Memory-Augmented In-context Reinforcement Learning (MAIC-RL) framework designed to improve CUDA optimization search capabilities of LLM-based GPU coding agents. KernelBlaster enables agents to learn from experience and make systematically informed decisions on future tasks by accumulating knowledge into a retrievable Persistent CUDA Knowledge Base. We propose a novel profile-guided, textual-gradient-based agentic flow for CUDA generation and optimization to achieve high performance across generations of GPU architectures. KernelBlaster guides LLM agents to systematically explore high-potential optimization strategies beyond naive rewrites. Compared to the PyTorch baseline, our method achieves geometric mean speedups of 1.43x, 2.50x, and 1.50x on KernelBench Levels 1, 2, and 3, respectively. We release KernelBlaster as an open-source agentic framework, accompanied by a test harness, verification components, and a reproducible evaluation pipeline.

In Diffusion Transformer (DiT) models, particularly for video generation, attention latency is a major bottleneck due to the long sequence length and the quadratic complexity. We find that attention weights can be separated into two parts: a small fraction of large weights with high rank and the remaining weights with very low rank. This naturally suggests applying sparse acceleration to the first part and low-rank acceleration to the second. Based on this finding, we propose SLA (Sparse-Linear Attention), a trainable attention method that fuses sparse and linear attention to accelerate diffusion models. SLA classifies attention weights into critical, marginal, and negligible categories, applying O(N^2) attention to critical weights, O(N) attention to marginal weights, and skipping negligible ones. SLA combines these computations into a single GPU kernel and supports both forward and backward passes. With only a few fine-tuning steps using SLA, DiT models achieve a 20x reduction in attention computation, resulting in significant acceleration without loss of generation quality. Experiments show that SLA reduces attention computation by 95% without degrading end-to-end generation quality, outperforming baseline methods. In addition, we implement an efficient GPU kernel for SLA, which yields a 13.7x speedup in attention computation and a 2.2x end-to-end speedup in video generation on Wan2.1-1.3B.

GPU architectures have continued to grow in complexity, with recent incarnations introducing increasingly powerful fixed-function units for matrix multiplication and data movement to accompany highly parallel general-purpose cores. To fully leverage these machines, software must use sophisticated schedules that maximally utilize all hardware resources. Since realizing such schedules is complex, both programmers and compilers routinely employ program transformations, such as software pipelining (SWP) and warp specialization (WS), to do so in practice. However, determining how best to use SWP and WS in combination is a challenging problem that is currently handled through a mix of brittle compilation heuristics and fallible human intuition, with little insight into the space of solutions. To remedy this situation, we introduce a novel formulation of SWP and WS as a joint optimization problem that can be solved holistically by off-the-shelf constraint solvers. We reify our approach in Twill, the first system that automatically derives optimal SWP and WS schedules for a large class of iterative programs. Twill is heuristic-free, easily extensible to new GPU architectures, and guaranteed to produce optimal schedules. We show that Twill can rediscover, and thereby prove optimal, the SWP and WS schedules manually developed by experts for Flash Attention on both the NVIDIA Hopper and Blackwell GPU architectures.

Fine-tuning large language models (LLMs) for downstream tasks can greatly improve model quality, however serving many different fine-tuned LLMs concurrently for users in multi-tenant environments is challenging. Dedicating GPU memory for each model is prohibitively expensive and naively swapping large model weights in and out of GPU memory is slow. Our key insight is that fine-tuned models can be quickly swapped in and out of GPU memory by extracting and compressing the delta between each model and its pre-trained base model. We propose DeltaZip, an LLM serving system that efficiently serves multiple full-parameter fine-tuned models concurrently by aggressively compressing model deltas by a factor of 6times to 8times while maintaining high model quality. DeltaZip increases serving throughput by 1.5times to 3times and improves SLO attainment compared to a vanilla HuggingFace serving system.

Scaling up Large Language Model(LLM) training involves fitting a tremendous amount of training parameters across a limited number of workers. However, methods like ZeRO-3 that drastically reduce GPU memory pressure often incur heavy communication to ensure global synchronization and consistency. Established efforts such as ZeRO++ use secondary partitions to avoid inter-node communications, given that intra-node GPU-GPU transfer generally has more bandwidth and lower latency than inter-node connections. However, as more capable infrastructure like Frontier, equipped with AMD GPUs, emerged with impressive computing capability, there is a need for investigations on the hardware topology and to develop targeted strategies to improve training efficiency. In this work, we propose a collection of communication and optimization strategies for ZeRO++ to reduce communication costs and improve memory utilization. In this paper, we propose a 3-level hierarchical partitioning specifically for the current Top-1 supercomputing cluster, Frontier, which aims at leveraging various bandwidths across layers of communications (GCD-GCD, GPU-GPU, and inter-node) to reduce communication overhead. For a 20B GPT model, we observe a 1.71x increase in TFLOPS per GPU when compared with ZeRO++ up to 384 GCDs and a scaling efficiency of 0.94 for up to 384 GCDs. To the best of our knowledge, our work is also the first effort to efficiently optimize LLM workloads on Frontier AMD GPUs.

Large language models (LLMs) have demonstrated strong code generation capabilities, yet the runtime performance of generated code is not guaranteed, and there have been few attempts to train LLMs using runtime performance as a reward in the HPC domain. We propose an online reinforcement learning approach that executes LLM-generated code on a supercomputer and directly feeds back the measured runtime performance (GFLOPS) as a reward. We further introduce a Staged Quality-Diversity (SQD) algorithm that progressively varies the permitted optimization techniques on a per-problem basis, enabling the model to learn code optimization from diverse perspectives. We build a distributed system connecting a GPU training cluster with a CPU benchmarking cluster, and train Qwen2.5 Coder 14B on a double-precision matrix multiplication task using Group Relative Policy Optimization (GRPO). Through two experiments, we show that reinforcement learning combining runtime performance feedback with staged optimization can improve the HPC code generation capability of LLMs.

Generative models are reshaping the live-streaming industry by redefining how content is created, styled, and delivered. Previous image-based streaming diffusion models have powered efficient and creative live streaming products but have hit limits on temporal consistency due to the foundation of image-based designs. Recent advances in video diffusion have markedly improved temporal consistency and sampling efficiency for offline generation. However, offline generation systems primarily optimize throughput by batching large workloads. In contrast, live online streaming operates under strict service-level objectives (SLOs): time-to-first-frame must be minimal, and every frame must meet a per-frame deadline with low jitter. Besides, scalable multi-GPU serving for real-time streams remains largely unresolved so far. To address this, we present StreamDiffusionV2, a training-free pipeline for interactive live streaming with video diffusion models. StreamDiffusionV2 integrates an SLO-aware batching scheduler and a block scheduler, together with a sink-token--guided rolling KV cache, a motion-aware noise controller, and other system-level optimizations. Moreover, we introduce a scalable pipeline orchestration that parallelizes the diffusion process across denoising steps and network layers, achieving near-linear FPS scaling without violating latency guarantees. The system scales seamlessly across heterogeneous GPU environments and supports flexible denoising steps (e.g., 1--4), enabling both ultra-low-latency and higher-quality modes. Without TensorRT or quantization, StreamDiffusionV2 renders the first frame within 0.5s and attains 58.28 FPS with a 14B-parameter model and 64.52 FPS with a 1.3B-parameter model on four H100 GPUs, making state-of-the-art generative live streaming practical and accessible--from individual creators to enterprise-scale platforms.

Native GPU kernel generation turns high-level tensor programs into executable, efficient low-level code. Existing Large Language Models (LLMs) struggle with this task, while execution-based reinforcement learning suffers from sparse rewards, reward hacking, and training instability. We present MusaCoder, a full-stack training framework for native GPU kernel generation on CUDA and MUSA backends. MusaCoder combines progressive kernel-oriented data synthesis, diversity-preserving rejection fine-tuning, and execution-feedback Reinforcement Learning (RL) through MooreEval, a distributed verifier and reward environment. To stabilize RL, MusaCoder introduces PrimeEcho for first-turn-anchored multi-turn rewards, Buffered Dynamic Retry for recovering signals from all-failed hard samples, and MirrorPop for off-policy sequence filtering. Experiments on KernelBench and a MUSA-ported variant show that MusaCoder outperforms strong open-source and proprietary baselines in both correctness and empirical speedup, with the 9B model matching or exceeding frontier closed-source models and the 27B model establishing a new state of the art. These results demonstrate not only the effectiveness of full-stack execution-feedback training for native kernel generation, but also the capability of Moore Threads GPUs to support the complete LLM post-training stack, providing a practical foundation for large-model training and optimization on emerging accelerators.

GPU kernels are the workhorse of modern deep learning, and optimizing them (via evolutionary search or coding agents) usually requires repeated measurement on target hardware. While these measurements provide the ground-truth signal necessary for kernel search, they are costly, because each evaluation of a kernel requires compilation and repeated execution on a GPU. As improvements in LLM inference reduce the cost of writing novel kernels and LLM-driven searches scale to large search budgets, on-device evaluation becomes a bottleneck. To address this, we study how LLMs can serve as selective GPU surrogates for kernel evaluation, by forecasting the performance of proposed kernels. A useful surrogate should be accurate, and it should be selective, by knowing when it could be wrong, and deferring to the GPU. To evaluate surrogates, we measure whether their forecasts are accurate, calibrated, and practically useful for recovering fast kernels under limited GPU-measurement budgets. Next, we study whether reinforcement learning can improve forecast accuracy and confidence calibration. Our experiments demonstrate that LLMs can accurately forecast relative kernel performance, that their utility can be improved through reinforcement learning. Used inside a kernel search, the surrogate lets the search consider several times as many candidates under the same GPU evaluation budget, and that leads to finding faster kernels than an equal-budget baseline. These results suggest that LLMs can play a broader role in kernel optimization, by acting as virtual models of a GPU rather than solely as kernel generators for search.

Fine-tuning large language models (LLMs) remains resource-intensive due to their sheer scale. While zeroth-order (ZO) optimization provides a memory-efficient alternative by eliminating backward passes, its application to multi-hundred-billion-parameter models is constrained by GPU memory and compute throughput. The ZO2 framework addresses the memory bottleneck by offloading model parameters to CPU memory and overlapping transformer block transfer with dual forward computation on a single GPU. However, ZO2 remains limited by its single-device execution and achieves modest throughput. In this work, we present DistZO2, a high-throughput, memory-efficient framework for distributed zeroth-order fine-tuning of LLMs. DistZO2 introduces three parallel strategies: (1) Perturbation Parallelism (PertP), which parallelizes the two perturbed forward passes across devices; (2) Distributed Data Parallelism (DDP), adapted to the scalar-gradient nature of ZO training; and (3) a unified 2D Parallelism design that combines PertP and DDP. To further mitigate communication bottlenecks introduced by parameter offloading, we propose a hardware-aware communication strategy that slices parameter blocks and redistributes them across GPUs via high-speed interconnects such as NVLink. DistZO2 scales zeroth-order fine-tuning to modern multi-GPU systems, preserving ZO2's memory efficiency while substantially improving training throughput. In our experiments on OPT-175B, DistZO2 achieves a 3x speedup over ZO2 with distributed computing. DistZO2's code has been open-sourced in https://github.com/liangyuwang/zo2.

This paper explores the problem of collision-free motion generation for manipulators by formulating it as a global motion optimization problem. We develop a parallel optimization technique to solve this problem and demonstrate its effectiveness on massively parallel GPUs. We show that combining simple optimization techniques with many parallel seeds leads to solving difficult motion generation problems within 50ms on average, 60x faster than state-of-the-art (SOTA) trajectory optimization methods. We achieve SOTA performance by combining L-BFGS step direction estimation with a novel parallel noisy line search scheme and a particle-based optimization solver. To further aid trajectory optimization, we develop a parallel geometric planner that plans within 20ms and also introduce a collision-free IK solver that can solve over 7000 queries/s. We package our contributions into a state of the art GPU accelerated motion generation library, cuRobo and release it to enrich the robotics community. Additional details are available at https://curobo.org

Recent advances in transformer-based foundation models have made them the default choice for many tasks, but their rapidly growing size makes fitting a full model on a single GPU increasingly difficult and their computational cost prohibitive. Block low-rank (BLR) compression techniques address this challenge by learning compact representations of weight matrices. While traditional low-rank (LR) methods often incur sharp accuracy drops, BLR approaches such as Monarch and BLAST can better capture the underlying structure, thus preserving accuracy while reducing computations and memory footprints. In this work, we use roofline analysis to show that, although BLR methods achieve theoretical savings and practical speedups for single-token inference, multi-token inference often becomes memory-bound in practice, increasing latency despite compiler-level optimizations in PyTorch. To address this, we introduce custom Triton kernels with partial fusion and memory layout optimizations for both Monarch and BLAST. On memory-constrained NVIDIA GPUs such as Jetson Orin Nano and A40, our kernels deliver up to 3.76times speedups and 3times model size compression over PyTorch dense baselines using CUDA backend and compiler-level optimizations, while supporting various models including Llama-7/1B, GPT2-S, DiT-XL/2, and ViT-B. Our code is available at https://github.com/pabillam/mem-efficient-blr.

AscendC (Ascend C) operator optimization on Huawei Ascend neural processing units (NPUs) faces a two-fold knowledge bottleneck: unlike the CUDA ecosystem, there are few public reference implementations to learn from, and performance hinges on a coupled two-part artifact - a host-side tiling program that orchestrates data movement and a kernel program that schedules and pipelines instructions. We present AscendOptimizer, an episodic agent that bootstraps this missing expertise by turning execution into experience. On the host side, AscendOptimizer performs profiling-in-the-loop evolutionary search to discover valid and high-performing tiling and data-movement configurations directly from hardware feedback. On the kernel side, it mines transferable optimization motifs by rewinding optimized kernels - systematically de-optimizing them to synthesize instructive "bad-to-good" trajectories - and distills these motifs into a retrievable experience bank for guided rewriting. By alternating host tuning and kernel rewriting in a closed loop, AscendOptimizer steadily expands feasibility and pushes latency down. On a benchmark of 127 real AscendC operators, AscendOptimizer achieves a 1.19x geometric-mean speedup over the open-source baseline, with 49.61% of operators outperforming their references, outperforming strong agent and search baselines.

Recent advances in large language models have brought immense value to the world, with their superior capabilities stemming from the massive number of parameters they utilize. However, even the GPUs with the highest memory capacities, currently peaking at 80GB, are far from sufficient to accommodate these vast parameters and their associated optimizer states when conducting stochastic gradient descent-based optimization. One approach to hosting such huge models is to aggregate device memory from many GPUs. However, this approach introduces prohibitive costs for most academic researchers, who always have a limited budget for many high-end GPU servers. In this paper, we focus on huge model fine-tuning on a single, even low-end, GPU in a commodity server, which is accessible to most AI researchers. In such a scenario, the state-of-the-art work ZeRO-Infinity suffers from two severe issues when running in a commodity server: 1) low GPU utilization due to inefficient swapping, and 2) limited trainable model size due to CPU memory capacity. The underlying reason is that ZeRO-Infinity is optimized for running on high-end GPU servers. To this end, we present Fuyou, a low-cost training framework that enables efficient 100B huge model fine-tuning on a low-end server with a low-end GPU and limited CPU memory capacity. The key idea is to add the SSD-CPU communication as an optimization dimension and thus carefully co-optimize computation and data swapping from a systematic approach to maximize GPU utilization. The experimental results show that 1) Fuyou is able to fine-tune 175B GPT-3 on a consumer GPU RTX 4090 with high GPU utilization, while ZeRO-Infinity fails to fine-tune; and 2) when training a small GPT-3 13B model, Fuyou achieves 156 TFLOPS on an RTX 4090 GPU while ZeRO-Infinity only achieves 45 TFLOPS.

The increasing complexity of machine learning models and the proliferation of diverse hardware architectures (CPUs, GPUs, accelerators) make achieving optimal performance a significant challenge. Heterogeneity in instruction sets, specialized kernel requirements for different data types and model features (e.g., sparsity, quantization), and architecture-specific optimizations complicate performance tuning. Manual optimization is resource-intensive, while existing automatic approaches often rely on complex hardware-specific heuristics and uninterpretable intermediate representations, hindering performance portability. We introduce PerfLLM, a novel automatic optimization methodology leveraging Large Language Models (LLMs) and Reinforcement Learning (RL). Central to this is PerfDojo, an environment framing optimization as an RL game using a human-readable, mathematically-inspired code representation that guarantees semantic validity through transformations. This allows effective optimization without prior hardware knowledge, facilitating both human analysis and RL agent training. We demonstrate PerfLLM's ability to achieve significant performance gains across diverse CPU (x86, Arm, RISC-V) and GPU architectures.

LLM-based agents for GPU kernel generation are advancing rapidly, yet their progress is fundamentally constrained by the benchmarks they optimize against. Existing benchmarks are poorly aligned with production inference frameworks: they evaluate kernels on a single GPU with synthetic inputs, ignore the surrounding compilation stack, and reward replicating known optimizations rather than discovering new ones. The resulting reward signals are misleading: agents learn to generate kernels that score well in sandboxes but introduce interface incompatibilities, compilation-stack conflicts, and silent correctness degradation when integrated into real systems. We introduce FastKernels, a kernel benchmark built around a minimal set of 46 representative architectures spanning 8 categories, whose kernels collectively subsume those of 96.2% (409/425) of HuggingFace Transformers architectures. FastKernels doubles as a minimalistic, production-grade inference framework that runs at parity with hardened systems such as vLLM and SGLang on mainstream LLM serving and substantially exceeds upstream references on under-served architectures; each task's interface mirrors the corresponding module in the state-of-the-art library for its architecture family, enabling direct deployment of optimized kernels into production codebases. Evaluating state-of-the-art kernel agents on FastKernels, we find that even the strongest agent achieves only 0.94times aggregate speedup over production baselines, with weaker agents at 0.78times and 0.53times -- confirming that benchmark-production misalignment is a critical bottleneck for the field. We release FastKernels as a stepping stone toward kernel agents whose benchmark gains translate directly into production throughput improvements. Code is available at https://github.com/Snowflake-AI-Research/fastkernels

Large language models (LLMs) such as GPT-3, OPT, and LLaMA have demonstrated remarkable accuracy in a wide range of tasks. However, training these models can incur significant expenses, often requiring tens of thousands of GPUs for months of continuous operation. Typically, this training is carried out in specialized GPU clusters equipped with homogeneous high-speed Remote Direct Memory Access (RDMA) network interface cards (NICs). The acquisition and maintenance of such dedicated clusters is challenging. Current LLM training frameworks, like Megatron-LM and Megatron-DeepSpeed, focus primarily on optimizing training within homogeneous cluster settings. In this paper, we introduce Holmes, a training framework for LLMs that employs thoughtfully crafted data and model parallelism strategies over the heterogeneous NIC environment. Our primary technical contribution lies in a novel scheduling method that intelligently allocates distinct computational tasklets in LLM training to specific groups of GPU devices based on the characteristics of their connected NICs. Furthermore, our proposed framework, utilizing pipeline parallel techniques, demonstrates scalability to multiple GPU clusters, even in scenarios without high-speed interconnects between nodes in distinct clusters. We conducted comprehensive experiments that involved various scenarios in the heterogeneous NIC environment. In most cases, our framework achieves performance levels close to those achievable with homogeneous RDMA-capable networks (InfiniBand or RoCE), significantly exceeding training efficiency within the pure Ethernet environment. Additionally, we verified that our framework outperforms other mainstream LLM frameworks under heterogeneous NIC environment in terms of training efficiency and can be seamlessly integrated with them.

Large language models (LLMs) have grown beyond the memory capacity of single GPU devices, necessitating quantization techniques for practical deployment. While NF4 (4-bit NormalFloat) quantization enables 4times memory reduction, inference on current NVIDIA GPUs (e.g., Ampere A100) requires expensive dequantization back to FP16 format, creating a critical performance bottleneck. This paper presents a lightweight shared memory optimization that addresses this gap through principled memory hierarchy exploitation while maintaining full ecosystem compatibility. We compare our technique against the open-source BitsAndBytes implementation, achieving 2.0--2.2times kernel speedup across three models (Gemma 27B, Qwen3 32B, and Llama3.3 70B) and up to 1.54times end-to-end improvement by leveraging the 12--15times latency advantage of shared memory over global memory access. Our optimization reduces instruction counts through simplified indexing logic while using only 64 bytes of shared memory per thread block, demonstrating that lightweight optimizations can deliver substantial performance gains with minimal engineering effort. This work provides a plug-and-play solution for the HuggingFace ecosystem that democratizes access to advanced models on existing GPU infrastructure.

The high memory and computation demand of large language models (LLMs) makes them challenging to be deployed on consumer devices due to limited GPU memory. Offloading can mitigate the memory constraint but often suffers from low GPU utilization, leading to low inference efficiency. In this work, we propose a novel framework, called pipelined offloading (PIPO), for efficient inference on consumer devices. PIPO designs a fine-grained offloading pipeline, complemented with optimized data transfer and computation, to achieve high concurrency and efficient scheduling for inference. Experimental results show that compared with state-of-the-art baseline, PIPO increases GPU utilization from below 40% to over 90% and achieves up to 3.1times higher throughput, running on a laptop equipped with a RTX3060 GPU of 6GB memory.

Existing large language model (LLM) serving systems fall into two categories: 1) a unified system where prefill phase and decode phase are co-located on the same GPU, sharing the unified computational resource and storage, and 2) a disaggregated system where the two phases are disaggregated to different GPUs. The design of the disaggregated system addresses the latency interference and sophisticated scheduling issues in the unified system but leads to storage challenges including 1) replicated weights for both phases that prevent flexible deployment, 2) KV cache transfer overhead between the two phases, 3) storage imbalance that causes substantial wasted space of the GPU capacity, and 4) suboptimal resource adjustment arising from the difficulties in migrating KV cache. Such storage inefficiency delivers poor serving performance under high request rates. In this paper, we identify that the advantage of the disaggregated system lies in the disaggregated computation, i.e., partitioning the computational resource to enable the asynchronous computation of two phases. Thus, we propose a novel LLM serving system, semi-PD, characterized by disaggregated computation and unified storage. In semi-PD, we introduce a computation resource controller to achieve disaggregated computation at the streaming multi-processor (SM) level, and a unified memory manager to manage the asynchronous memory access from both phases. semi-PD has a low-overhead resource adjustment mechanism between the two phases, and a service-level objective (SLO) aware dynamic partitioning algorithm to optimize the SLO attainment. Compared to state-of-the-art systems, semi-PD maintains lower latency at higher request rates, reducing the average end-to-end latency per request by 1.27-2.58x on DeepSeek series models, and serves 1.55-1.72x more requests adhering to latency constraints on Llama series models.

The rapid growth of deep learning has driven exponential increases in model parameters and computational demands. NVIDIA GPUs and their CUDA-based software ecosystem provide robust support for parallel computing, significantly alleviating computational bottlenecks. Meanwhile, due to the cultivation of user programming habits and the high performance of GPUs, the CUDA ecosystem has established a dominant position in the field of parallel software. This dominance requires other hardware platforms to support CUDA-based software with performance portability. However, translating CUDA code to other platforms poses significant challenges due to differences in parallel programming paradigms and hardware architectures. Existing approaches rely on language extensions, domain-specific languages (DSLs), or compilers but face limitations in workload coverage and generalizability. Moreover, these methods often incur substantial development costs. Recently, LLMs have demonstrated extraordinary potential in various vertical domains, especially in code-related tasks. However, the performance of existing LLMs in CUDA transpilation, particularly for high-performance code, remains suboptimal. To address these challenges, we propose a novel framework for generating high-performance CUDA and corresponding platform code pairs, leveraging AI compiler and automatic optimization technology. We further enhance the framework with a graph-based data augmentation method and introduce HPCTransEval, a benchmark for evaluating LLM performance on CUDA transpilation. We conduct experiments using CUDA-to-CPU transpilation as a case study on leading LLMs. The speedup ratio of the CPU operators has an average improvemnet of 43.8\%, highlighting the potential of LLMs to address compatibility challenges within the CUDA ecosystem. Our code is available at https://github.com/PJLAB-CHIP/HPCTransCompile.

High-performance GPU kernel optimization remains a critical yet labor-intensive task in modern machine learning workloads. Although Triton, a domain-specific language for GPU programming, enables developers to write efficient kernels with concise code, achieving expert-level performance still requires deep understanding of GPU architectures and low-level performance trade-offs. We present TritonForge, a profiling-guided framework for automated Triton kernel optimization. TritonForge integrates kernel analysis, runtime profiling, and iterative code transformation to streamline the optimization process. By incorporating feedback from profiling results, the system identifies performance bottlenecks, proposes targeted code modifications, and evaluates their impact automatically. Across diverse kernel types, TritonForge achieves up to 5x performance improvement over baseline implementations and on average 1.76x of the cases are successful, providing a foundation for future research in automated GPU performance optimization.

While the rapid expansion of data centers poses challenges for power grids, it also offers new opportunities as potentially flexible loads. Existing power system research often abstracts data centers as aggregate resources, while computer system research primarily focuses on optimizing GPU energy efficiency and largely ignores the grid impacts of optimized GPU power consumption. To bridge this gap, we develop a GPU-to-Grid framework that couples device-level GPU control with power system objectives. We study distribution-level voltage regulation enabled by flexibility in LLM inference, using batch size as a control knob that trades off the voltage impacts of GPU power consumption against inference latency and token throughput. We first formulate this problem as an optimization problem and then realize it as an online feedback optimization controller that leverages measurements from both the power grid and GPU systems. Our key insight is that reducing GPU power consumption alleviates violations of lower voltage limits, while increasing GPU power mitigates violations near upper voltage limits in distribution systems; this runs counter to the common belief that minimizing GPU power consumption is always beneficial to power grids.

Deploying large language models (LLMs) for online inference is often constrained by limited GPU memory, particularly due to the growing KV cache during auto-regressive decoding. Hybrid GPU-CPU execution has emerged as a promising solution by offloading KV cache management and parts of attention computation to the CPU. However, a key bottleneck remains: existing schedulers fail to effectively overlap CPU-offloaded tasks with GPU execution during the latency-critical, bandwidth-bound decode phase. This particularly penalizes real-time, decode-heavy applications (e.g., chat, Chain-of-Thought reasoning) which are currently underserved by existing systems, especially under memory pressure typical of edge or low-cost deployments. We present APEX, a novel, profiling-informed scheduling strategy that maximizes CPU-GPU parallelism during hybrid LLM inference. Unlike systems relying on static rules or purely heuristic approaches, APEX dynamically dispatches compute across heterogeneous resources by predicting execution times of CPU and GPU subtasks to maximize overlap while avoiding scheduling overheads. We evaluate APEX on diverse workloads and GPU architectures (NVIDIA T4, A10), using LLaMa-2-7B and LLaMa-3.1-8B models. Compared to GPU-only schedulers like VLLM, APEX improves throughput by 84% - 96% on T4 and 11% - 89% on A10 GPUs, while preserving latency. Against the best existing hybrid schedulers, it delivers up to 49% (T4) and 37% (A10) higher throughput in long-output settings. APEX significantly advances hybrid LLM inference efficiency on such memory-constrained hardware and provides a blueprint for scheduling in heterogeneous AI systems, filling a critical gap for efficient real-time LLM applications.

Training transformer models requires substantial GPU compute and memory resources. In homogeneous clusters, distributed strategies allocate resources evenly, but this approach is inefficient for heterogeneous clusters, where GPUs differ in power and memory. As high-end GPUs are costly and limited in availability, heterogeneous clusters with diverse GPU types are becoming more common. Existing methods attempt to balance compute across GPUs based on capacity but often underutilize compute due to memory constraints. We present Cephalo, a system that optimizes compute and memory usage by decoupling compute distribution from training state assignment. Cephalo outperforms state-of-the-art methods by achieving significantly higher training throughput while supporting larger models and batch sizes.

Developing efficient GPU kernels is essential for scaling modern AI systems, yet it remains a complex task due to intricate hardware architectures and the need for specialized optimization expertise. Although Large Language Models (LLMs) demonstrate strong capabilities in general sequential code generation, they face significant challenges in GPU code generation because of the scarcity of high-quality labeled training data, compiler biases when generating synthetic solutions, and limited generalization across hardware generations. This precludes supervised fine-tuning (SFT) as a scalable methodology for improving current LLMs. In contrast, reinforcement learning (RL) offers a data-efficient and adaptive alternative but requires access to relevant tools, careful selection of training problems, and a robust evaluation environment. We present Makora's environment and tools for reinforcement learning finetuning of frontier models and report our results from fine-tuning GPT-5 for Triton code generation. In the single-attempt setting, our fine-tuned model improves kernel correctness from 43.7% to 77.0% (+33.3 percentage points) and increases the fraction of problems outperforming TorchInductor from 14.8% to 21.8% (+7 percentage points) compared to baseline GPT-5, while exceeding prior state-of-the-art models on KernelBench. When integrated into a full coding agent, it is able to solve up to 97.4% of problems in an expanded KernelBench suite, outperforming the PyTorch TorchInductor compiler on 72.9% of problems with a geometric mean speedup of 2.12x. Our work demonstrates that targeted post-training with reinforcement learning can unlock LLM capabilities in highly specialized technical domains where traditional supervised learning is limited by data availability, opening new pathways for AI-assisted accelerator programming.

Developing efficient CUDA kernels is increasingly critical for AI applications such as large-scale LLM training. However, manual kernel design is both costly and time-consuming, motivating automatic approaches that leverage LLMs for code generation. Existing methods for automatic kernel generation, however, often produce low-efficiency kernels, incur high computational overhead, and fail to generalize across settings. In this work, we propose CudaForge, a training-free multi-agent workflow for CUDA kernel generation and optimization. Our workflow is inspired by the iterative workflow of human experts, which contains steps such as developing initial kernels, testing correctness, analyzing hardware feedback, and iterative improvement. More specifically, CudaForge employs two LLM agents: a Coder and a Judge, that iteratively generate, correct, and optimize CUDA kernels, while integrating hardware feedback such as Nsight Compute (NCU) metrics. In extensive evaluations, we show that CudaForge, by leveraging base models like OpenAI-o3, achieves 97.6\% correctness of generated kernels and an average 1.68times speedup over PyTorch baselines, substantially surpassing state-of-the-art models including OpenAI-o3 and Kevin on KernelBench.Beyond accuracy and speed, CudaForge demonstrates strong generalization across GPUs (A100, RTX 6000, 4090, 3090) and base models (OpenAI-o3, GPT-5, gpt-oss-120B, Claude-Sonnet-4, QwQ-32B), while maintaining high efficiency. In particular, generating an optimized kernel takes about 26.5 minutes on one RTX6000 and incurs about \ 0.3 API cost, which is significantly cheaper than existing agentic work that costs 6 H100 hours and 5 API cost per kernel. Our results highlight that multi-agent, training-free workflows can enable cost-effective, generalizable, and high-performance CUDA kernel optimization. Code available at https://github.com/OptimAI-Lab/CudaForge

Existing GPU libraries often struggle to fully exploit the parallel resources and on-chip memory (SRAM) of GPUs when chaining multiple GPU functions as individual kernels. While Kernel Fusion (KF) techniques like Horizontal Fusion (HF) and Vertical Fusion (VF) can mitigate this, current library implementations often require library developers to manually create fused kernels. Hence, library users rely on limited sets of pre-compiled or template-based fused kernels. This limits the use cases that can benefit from HF and VF and increases development costs. In order to solve these issues, we present a novel methodology for building GPU libraries that enables automatic on-demand HF and VF for arbitrary combinations of GPU library functions. Our methodology defines reusable, fusionable components that users combine via high-level programming interfaces. Leveraging C++17 metaprogramming features available in compilers like nvcc, our methodology generates a single and optimized fused kernel tailored to the user's specific sequence of operations at compile time, without needing a custom compiler or manual development and pre-compilation of kernel combinations. This approach abstracts low-level GPU complexities while maximizing GPU resource utilization and keeping intermediate data in SRAM. We provide an open-source implementation demonstrating significant speedups compared to traditional libraries in various benchmarks, validating the effectiveness of this methodology for improving GPU performance in the range of 2x to more than 1000x, while preserving high-level programmability.

Large language model (LLM)-based inference workloads increasingly dominate data center costs and resource utilization. Therefore, understanding the inference workload characteristics on evolving CPU-GPU coupled architectures is crucial for optimization. This paper presents an in-depth analysis of LLM inference behavior on loosely-coupled (PCIe A100/H100) and closely-coupled (GH200) systems. We analyze performance dynamics using fine-grained operator-to-kernel trace analysis, facilitated by our novel profiler SKIP and metrics like Total Kernel Launch and Queuing Time (TKLQT). Results show that closely-coupled (CC) GH200 significantly outperforms loosely-coupled (LC) systems at large batch sizes, achieving 1.9x-2.7x faster prefill latency for Llama 3.2-1B. However, our analysis also reveals that GH200 remains CPU-bound up to 4x larger batch sizes than LC systems. In this extended CPU-bound region, we identify the performance characteristics of the Grace CPU as a key factor contributing to higher inference latency at low batch sizes on GH200. We demonstrate that TKLQT accurately identifies this CPU/GPU-bound transition point. Based on this analysis, we further show that kernel fusion offers significant potential to mitigate GH200's low-batch latency bottleneck by reducing kernel launch overhead. This detailed kernel-level characterization provides critical insights for optimizing diverse CPU-GPU coupling strategies. This work is an initial effort, and we plan to explore other major AI/DL workloads that demand different degrees of CPU-GPU heterogeneous architectures.

GPU kernel generation by LLMs has recently experienced rapid development, leveraging test-time scaling and reinforcement learning techniques. However, a key challenge for kernel generation is the scarcity of high-quality data, as most high-quality kernels are proprietary and not open-source. This challenge prevents us from leveraging supervised fine-tuning to align LLMs to the kernel generation task. To address this challenge, we develop a pipeline that generates and curates high-quality CUDA kernels with reasoning traces, motivated by a critical observation that concise yet informative reasoning traces result in robust generation of high-performance kernels. Using this pipeline, we construct our dataset ConCuR and introduce our model KernelCoder, which is the first model trained on a curated dataset consisting of PyTorch, reasoning, and CUDA kernel pairs, to our knowledge. In the KernelBench setup, our model achieves significant improvements over the existing top-performing model, QwQ-32B, and outperforms all open-source models fine-tuned for kernel generation, as well as frontier models such as DeepSeek-V3.1-Think and Claude-4-sonnet. Finally, we show that the average reasoning length can serve as a metric to assess the difficulty of kernel generation tasks. The observations, metrics, and our data collection and curation pipeline can help obtain better data in the kernel generation task in the future.

The rapid adoption of large language models (LLMs) has created significant challenges for efficient inference at scale. Unlike traditional workloads, LLM inference is constrained by both computation and the memory overhead of key-value (KV) caching, which accelerates decoding but quickly exhausts GPU memory. In this paper, we introduce the first queueing-theoretic framework that explicitly incorporates both computation and GPU memory constraints into the analysis of LLM inference. Based on this framework, we derive rigorous stability and instability conditions that determine whether an LLM inference service can sustain incoming demand without unbounded queue growth. This result offers a powerful tool for system deployment, potentially addressing the core challenge of GPU provisioning. By combining an estimated request arrival rate with our derived stable service rate, operators can calculate the necessary cluster size to avoid both costly over-purchasing and performance-violating under-provisioning. We further validate our theoretical predictions through extensive experiments in real GPU production environments. Our results show that the predicted stability conditions are highly accurate, with deviations typically within 10%.

The computational sparsity of Mixture-of-Experts (MoE) models enables sub-linear growth in compute cost as model size increases, thus offering a scalable path to training massive neural networks. However, existing implementations suffer from low GPU utilization, significant latency overhead, and a fundamental inability to leverage task locality, primarily due to CPU-managed scheduling, host-initiated communication, and frequent kernel launches. To overcome these limitations, we develop FlashMoE, a fully GPU-resident MoE operator that fuses expert computation and inter-GPU communication into a single persistent GPU kernel. FlashMoE enables fine-grained pipelining of dispatch, compute, and combine phases, eliminating launch overheads and reducing idle gaps. Unlike existing work, FlashMoE eliminates bulk-synchronous collectives for one-sided, device-initiated, inter-GPU (R)DMA transfers, thereby unlocking payload efficiency by eliminating bloated or redundant network payloads in sparsely activated layers. When evaluated on an 8-H100 GPU node with MoE models comprising up to 128 experts and 16K token sequences, FlashMoE achieves up to 9x higher GPU utilization, 6x lower latency, 5.7x higher throughput, and 4x better overlap efficiency compared to state-of-the-art baselines, despite using FP32, whereas the baselines use FP16. FlashMoE shows that principled GPU kernel-hardware co-design is key to unlocking the performance ceiling of large-scale distributed ML. We provide code at https://github.com/osayamenja/FlashMoE.

The rapid advancement of Large Language Models (LLMs) has driven the need for more efficient serving strategies. In this context, efficiency refers to the proportion of requests that meet their Service Level Objectives (SLOs), particularly for Time To First Token (TTFT) and Time Between Tokens (TBT). However, existing systems often prioritize one metric at the cost of the other. We present Ascendra, an LLM serving system designed to meet both TTFT and TBT SLOs simultaneously. The core insight behind Ascendra is that a request's urgency evolves as it approaches its deadline. To leverage this, Ascendra partitions GPU resources into two types of instances: low-priority and high-priority. Low-priority instances maximize throughput by processing requests out of arrival order, but at the risk of request starvation. To address this, Ascendra employs a performance model to predict requests at risk of missing their SLOs and proactively offloads them to high-priority instances. High-priority instances are optimized for low-latency execution and handle urgent requests nearing their deadlines. This partitioned architecture enables Ascendra to effectively balance high throughput and low latency. Extensive evaluation shows that Ascendra improves system throughput by up to 1.7x compared to vLLM and Sarathi-Serve while meeting both TTFT and TBT SLOs.

Progress in LLMs is increasingly measured through standardized benchmarks, where state-of-the-art improvements are often separated by fractions of a percentage point. At the same time, the computational cost of evaluating modern LLMs has driven widespread adoption of specialized inference backends, software systems that execute trained models efficiently at inference time. While critical for scalability, system-level optimizations, such as custom CUDA kernels and reduced-precision arithmetic, can alter token probabilities and introduce non-determinism, possibly cascading into divergent generation. In this work, we first survey the inference landscape, identifying 200 distinct engines, and analyze 35,000 ML publications, finding that the specific inference stack is rarely reported despite this widespread diversity. We then present a systematic empirical study of how inference backends affect LLM benchmark results. Holding model weights, decoding parameters, and hardware constant, we evaluate five widely used inference engines, including vLLM, SGLang, and llama.cpp, across multiple open-weight models and established benchmarks. We show that the choice of backend alone can shift benchmark scores by up to 16.6 percentage points and induce high rates of output disagreement. By isolating backend optimizations and tracing the execution pipeline, we find this divergence is driven by system-level optimizations like prefix caching and CUDA graphs, custom kernels, and engine-specific defaults in logit processing. Our findings identify the inference backend as a previously unreported but consequential hyperparameter in the evaluation of LLM and advocate standardized reporting of inference stacks to improve the reproducibility and interpretability of benchmark comparisons.

Writing high-performance GPU kernels is among the most labor-intensive tasks in machine learning systems engineering. We present AutoKernel, an open-source framework that applies an autonomous agent loop to GPU kernel optimization for arbitrary PyTorch models. Given a model, AutoKernel profiles it to identify computational bottlenecks, ranks them by Amdahl's law impact, and iteratively refines Triton or CUDA C++ kernel implementations through hundreds of experiments without human intervention. A five-stage correctness harness covering smoke tests, shape sweeps, numerical stability, determinism verification, and edge-case coverage ensures every candidate kernel is validated before any speedup is recorded. The system comprises over 9,000 lines of Python, 18 starter kernel implementations across two backends, a six-tier optimization playbook, and integration with the KernelBench benchmark suite. AutoKernel covers nine kernel types spanning the dominant operations in modern transformer architectures. On an NVIDIA H100, our Triton kernels outperform both PyTorch eager and torch.compile (max-autotune) on the majority of tested configurations: 5.29x over eager on RMSNorm, 2.82x on softmax, and 2.21x on cross-entropy, while beating torch.compile by 2.83x, 3.44x, and 2.94x respectively. In community deployment, an AutoKernel-optimized kernel achieved first place on the vectorsum_v2 B200 leaderboard. The full system is available at https://github.com/RightNow-AI/autokernel.

Agentic Variation Operators (AVO) are a new family of evolutionary variation operators that replace the fixed mutation, crossover, and hand-designed heuristics of classical evolutionary search with autonomous coding agents. Rather than confining a language model to candidate generation within a prescribed pipeline, AVO instantiates variation as a self-directed agent loop that can consult the current lineage, a domain-specific knowledge base, and execution feedback to propose, repair, critique, and verify implementation edits. We evaluate AVO on attention, among the most aggressively optimized kernel targets in AI, on NVIDIA Blackwell (B200) GPUs. Over 7 days of continuous autonomous evolution on multi-head attention, AVO discovers kernels that outperform cuDNN by up to 3.5% and FlashAttention-4 by up to 10.5% across the evaluated configurations. The discovered optimizations transfer readily to grouped-query attention, requiring only 30 minutes of additional autonomous adaptation and yielding gains of up to 7.0% over cuDNN and 9.3% over FlashAttention-4. Together, these results show that agentic variation operators move beyond prior LLM-in-the-loop evolutionary pipelines by elevating the agent from candidate generator to variation operator, and can discover performance-critical micro-architectural optimizations that produce kernels surpassing state-of-the-art expert-engineered attention implementations on today's most advanced GPU hardware.

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

We present 3DGS-LM, a new method that accelerates the reconstruction of 3D Gaussian Splatting (3DGS) by replacing its ADAM optimizer with a tailored Levenberg-Marquardt (LM). Existing methods reduce the optimization time by decreasing the number of Gaussians or by improving the implementation of the differentiable rasterizer. However, they still rely on the ADAM optimizer to fit Gaussian parameters of a scene in thousands of iterations, which can take up to an hour. To this end, we change the optimizer to LM that runs in conjunction with the 3DGS differentiable rasterizer. For efficient GPU parallization, we propose a caching data structure for intermediate gradients that allows us to efficiently calculate Jacobian-vector products in custom CUDA kernels. In every LM iteration, we calculate update directions from multiple image subsets using these kernels and combine them in a weighted mean. Overall, our method is 30% faster than the original 3DGS while obtaining the same reconstruction quality. Our optimization is also agnostic to other methods that acclerate 3DGS, thus enabling even faster speedups compared to vanilla 3DGS.

Serving Large Language Models (LLMs) is critical for AI-powered applications but demands substantial computational resources, particularly in memory bandwidth and computational throughput. Low-precision computation has emerged as a key technique to improve efficiency while reducing resource consumption. Existing approaches for generating low-precision kernels are limited to weight bit widths that are powers of two and suffer from suboptimal performance due to high-level GPU programming abstractions. These abstractions restrict critical optimizations, such as fine-grained register management and optimized memory access patterns, which are essential for efficient low-precision computations. In this paper, we introduce a virtual machine (VM) designed for General-Purpose GPU (GPGPU) computing, enabling support for low-precision data types with arbitrary bit widths while maintaining GPU programmability. The proposed VM features a thread-block-level programming model, a hierarchical memory space, a novel algebraic layout system, and extensive support for diverse low-precision data types. VM programs are compiled into highly efficient GPU programs with automatic vectorization and instruction selection. Extensive experiments demonstrate that our VM efficiently supports a full spectrum of low-precision data types, and outperforms state-of-the-art low-precision kernels on their supported types. Compared to existing compilers like Triton and Ladder, as well as hand-optimized kernels such as QuantLLM and Marlin, our VM achieves performance improvements of 1.75x, 2.61x, 1.29x and 1.03x, respectively.

AI agents are emerging as a dominant workload in a wide range of applications, promising to be the vehicle that delivers the promised benefits of AI to enterprises and consumers. Unlike conventional software or static inference, agentic workloads are dynamic and structurally complex. Often these agents are directed graphs of compute and IO operations that span multi-modal data input and conversion), data processing and context gathering (e.g vector DB lookups), multiple LLM inferences, tool calls, etc. To scale AI agent usage, we need efficient and scalable deployment and agent-serving infrastructure. To tackle this challenge, in this paper, we present a system design for dynamic orchestration of AI agent workloads on heterogeneous compute infrastructure spanning CPUs and accelerators, both from different vendors and across different performance tiers within a single vendor. The system delivers several building blocks: a framework for planning and optimizing agentic AI execution graphs using cost models that account for compute, memory, and bandwidth constraints of different HW; a MLIR based representation and compilation system that can decompose AI agent execution graphs into granular operators and generate code for different HW options; and a dynamic orchestration system that can place the granular components across a heterogeneous compute infrastructure and stitch them together while meeting an end-to-end SLA. Our design performs a systems level TCO optimization and preliminary results show that leveraging a heterogeneous infrastructure can deliver significant TCO benefits. A preliminary surprising finding is that for some workloads a heterogeneous combination of older generation GPUs with newer accelerators can deliver similar TCO as the latest generation homogenous GPU infrastructure design, potentially extending the life of deployed infrastructure.

GPU kernels are critical for ML performance but difficult to optimize across diverse accelerators. We present KForge, a platform-agnostic framework built on two collaborative LLM-based agents: a generation agent that produces and iteratively refines programs through compilation and correctness feedback, and a performance analysis agent that interprets profiling data to guide optimization. This agent-based architecture requires only a single-shot example to target new platforms. We make three key contributions: (1) introducing an iterative refinement system where the generation agent and performance analysis agent collaborate through functional and optimization passes, interpreting diverse profiling data (from programmatic APIs to GUI-based tools) to generate actionable recommendations that guide program synthesis for arbitrary accelerators; (2) demonstrating that the generation agent effectively leverages cross-platform knowledge transfer, where a reference implementation from one architecture substantially improves generation quality for different hardware targets; and (3) validating the platform-agnostic nature of our approach by demonstrating effective program synthesis across fundamentally different parallel computing platforms: NVIDIA CUDA and Apple Metal.

Differentiable neural architecture search methods became popular in recent years, mainly due to their low search costs and flexibility in designing the search space. However, these methods suffer the difficulty in optimizing network, so that the searched network is often unfriendly to hardware. This paper deals with this problem by adding a differentiable latency loss term into optimization, so that the search process can tradeoff between accuracy and latency with a balancing coefficient. The core of latency prediction is to encode each network architecture and feed it into a multi-layer regressor, with the training data which can be easily collected from randomly sampling a number of architectures and evaluating them on the hardware. We evaluate our approach on NVIDIA Tesla-P100 GPUs. With 100K sampled architectures (requiring a few hours), the latency prediction module arrives at a relative error of lower than 10%. Equipped with this module, the search method can reduce the latency by 20% meanwhile preserving the accuracy. Our approach also enjoys the ability of being transplanted to a wide range of hardware platforms with very few efforts, or being used to optimizing other non-differentiable factors such as power consumption.

Zeroth-order optimization (ZO) is a memory-efficient strategy for fine-tuning Large Language Models using only forward passes. However, the application of ZO fine-tuning in memory-constrained settings such as mobile phones and laptops is still challenging since full precision forward passes are infeasible. In this study, we address this limitation by integrating sparsity and quantization into ZO fine-tuning of LLMs. Specifically, we investigate the feasibility of fine-tuning an extremely small subset of LLM parameters using ZO. This approach allows the majority of un-tuned parameters to be quantized to accommodate the constraint of limited device memory. Our findings reveal that the pre-training process can identify a set of "sensitive parameters" that can guide the ZO fine-tuning of LLMs on downstream tasks. Our results demonstrate that fine-tuning 0.1% sensitive parameters in the LLM with ZO can outperform the full ZO fine-tuning performance, while offering wall-clock time speedup. Additionally, we show that ZO fine-tuning targeting these 0.1% sensitive parameters, combined with 4 bit quantization, enables efficient ZO fine-tuning of an Llama2-7B model on a GPU device with less than 8 GiB of memory and notably reduced latency.

This paper presents a novel approach, named the Group Marching Tree (GMT*) algorithm, to planning on GPUs at rates amenable to application within control loops, allowing planning in real-world settings via repeated computation of near-optimal plans. GMT*, like the Fast Marching Tree (FMT) algorithm, explores the state space with a "lazy" dynamic programming recursion on a set of samples to grow a tree of near-optimal paths. GMT*, however, alters the approach of FMT with approximate dynamic programming by expanding, in parallel, the group of all active samples with cost below an increasing threshold, rather than only the minimum cost sample. This group approximation enables low-level parallelism over the sample set and removes the need for sequential data structures, while the "lazy" collision checking limits thread divergence---all contributing to a very efficient GPU implementation. While this approach incurs some suboptimality, we prove that GMT* remains asymptotically optimal up to a constant multiplicative factor. We show solutions for complex planning problems under differential constraints can be found in ~10 ms on a desktop GPU and ~30 ms on an embedded GPU, representing a significant speed up over the state of the art, with only small losses in performance. Finally, we present a scenario demonstrating the efficacy of planning within the control loop (~100 Hz) towards operating in dynamic, uncertain settings.

Mixture of Experts (MoE) models have enabled the scaling of Large Language Models (LLMs) and Vision Language Models (VLMs) by achieving massive parameter counts while maintaining computational efficiency. However, MoEs introduce several inference-time challenges, including load imbalance across experts and the additional routing computational overhead. To address these challenges and fully harness the benefits of MoE, a systematic evaluation of hardware acceleration techniques is essential. We present MoE-Inference-Bench, a comprehensive study to evaluate MoE performance across diverse scenarios. We analyze the impact of batch size, sequence length, and critical MoE hyperparameters such as FFN dimensions and number of experts on throughput. We evaluate several optimization techniques on Nvidia H100 GPUs, including pruning, Fused MoE operations, speculative decoding, quantization, and various parallelization strategies. Our evaluation includes MoEs from the Mixtral, DeepSeek, OLMoE and Qwen families. The results reveal performance differences across configurations and provide insights for the efficient deployment of MoEs.

Can a single LLM-based optimization system match specialized tools across fundamentally different domains? We show that when optimization problems are formulated as improving a text artifact evaluated by a scoring function, a single AI-based optimization system-supporting single-task search, multi-task search with cross-problem transfer, and generalization to unseen inputs-achieves state-of-the-art results across six diverse tasks. Our system discovers agent architectures that nearly triple Gemini Flash's ARC-AGI accuracy (32.5% to 89.5%), finds scheduling algorithms that cut cloud costs by 40%, generates CUDA kernels where 87% match or beat PyTorch, and outperforms AlphaEvolve's reported circle packing solution (n=26). Ablations across three domains reveal that actionable side information yields faster convergence and substantially higher final scores than score-only feedback, and that multi-task search outperforms independent optimization given equivalent per-problem budget through cross-task transfer, with benefits scaling with the number of related tasks. Together, we show for the first time that text optimization with LLM-based search is a general-purpose problem-solving paradigm, unifying tasks traditionally requiring domain-specific algorithms under a single framework. We open-source optimize\_anything with support for multiple backends as part of the GEPA project at https://github.com/gepa-ai/gepa .

The challenge of mapping AI architectures to GPU hardware is creating a critical bottleneck in AI progress. Despite substantial efforts, hand-written custom kernels fail to meet their theoretical performance thresholds, even on well-established operations like linear attention. The diverse hardware capabilities of GPUs might suggest that we need a wide variety of techniques to achieve high performance. However, our work explores whether a small number of key abstractions can drastically simplify the process. We present ThunderKittens (TK), a framework for writing performant AI kernels while remaining easy to use and maintain. Our abstractions map to the three levels of the GPU hierarchy: (1) at the warp-level, we provide 16x16 matrix tiles as basic data structures and PyTorch-like parallel compute operations over tiles, (2) at the thread-block level, we provide a template for overlapping asynchronous operations across parallel warps, and (3) at the grid-level, we provide support to help hide the block launch and tear-down, and memory costs. We show the value of TK by providing kernels that match or outperform prior kernels for a range of AI operations. We match CuBLAS and FlashAttention-3 on GEMM and attention inference performance and outperform the strongest baselines by 10-40% on attention backwards, 8times on state space models, and 14times on linear attention.

Deploying a large language model (LLM) inference service remains costly because centralized serving depends on specialized GPU clusters and high-bandwidth interconnects in datacenters. An appealing alternative is to leverage collaborative decentralized GPU pools. However, heterogeneity in GPU and limited interconnected network bandwidth, along with potentially dynamic availability, make efficient scheduling the central challenge in this scenario. In this paper, we present Parallax, a decentralized LLM serving system that turns a pool of heterogeneous GPUs into an efficient inference platform via a two-phase scheduler. Parallax decomposes planning into (i) model allocation, which places layers of each replica across diverse GPUs to jointly optimize latency and throughput under memory and link-bandwidth constraints, and (ii) request-time GPU pipeline selection, which stitches layers from different replicas into end-to-end execution chains that balance load and adapt to current conditions. We implement Parallax and evaluate it on open-source LLMs deployed over real volunteer nodes. Parallax consistently reduces latency and increases throughput relative to decentralized baselines, demonstrating that principled scheduling can make volunteer compute a practical, affordable substrate for LLM inference. Github Repo at: https://github.com/GradientHQ/parallax.

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

As the size of large language models grows exponentially, GPU memory has become a bottleneck for adapting these models to downstream tasks. In this paper, we aim to push the limits of memory-efficient training by minimizing memory usage on model weights, gradients, and optimizer states, within a unified framework. Our idea is to eliminate both gradients and optimizer states using zeroth-order optimization, which approximates gradients by perturbing weights during forward passes to identify gradient directions. To minimize memory usage on weights, we employ model quantization, e.g., converting from bfloat16 to int4. However, directly applying zeroth-order optimization to quantized weights is infeasible due to the precision gap between discrete weights and continuous gradients, which would otherwise require de-quantization and re-quantization. To overcome this challenge, we propose Quantized Zeroth-order Optimization (QZO), a novel approach that perturbs the continuous quantization scale for gradient estimation and uses a directional derivative clipping method to stabilize training. QZO is orthogonal to both scalar-based and codebook-based post-training quantization methods. Compared to full-parameter fine-tuning in bfloat16, QZO can reduce the total memory cost by more than 18times for 4-bit LLMs, and enables fine-tuning Llama-2-13B and Stable Diffusion 3.5 Large within a single 24GB GPU.

GPU kernel optimization is increasingly critical for efficient deep learning systems, but writing high-performance kernels still requires substantial low-level expertise. Recent AI coding agents can iteratively read code, invoke compilers and profilers, and refine implementations, yet existing kernel benchmarks evaluate single LLM calls rather than full agent workflows, and none include both kernel-to-kernel optimization and unseen-configuration generalization testing. We present AgentKernelArena, an open-source benchmark for measuring AI coding agents on GPU kernel optimization. The benchmark contains 196 tasks spanning HIP-to-HIP optimization, Triton-to-Triton optimization, and PyTorch-to-HIP translation, and evaluates complete agent workflows in isolated workspaces using gated compilation, correctness, and performance checks, centralized scoring and an unseen-configuration generalization protocol that tests whether optimizations transfer to input configurations the agent never observed. Across production agents including Cursor Agent, Claude Code, and Codex Agent, we find near-perfect compilation and high correctness rates on most task categories, with the strongest configurations achieving mean speedups of up to 6.89x on PyTorch-to-HIP, 6.69x on HIP-to-HIP, and 2.13x on Triton-to-Triton tasks. Our unseen-configuration evaluation shows that HIP-to-HIP and Triton-to-Triton optimizations largely transfer to unseen input shapes, while PyTorch-to-HIP exhibits substantial correctness drops, indicating that agents generating kernels from scratch frequently hardcode shape-specific assumptions. AgentKernelArena is designed as a modular, extensible framework for rigorous evaluation of agentic GPU kernel optimization across agents, tasks, and hardware targets.

k-means has historically been positioned primarily as an offline processing primitive, typically used for dataset organization or embedding preprocessing rather than as a first-class component in online systems. In this work, we revisit this classical algorithm under the lens of modern AI system design and enable k-means as an online primitive. We point out that existing GPU implementations of k-means remain fundamentally bottlenecked by low-level system constraints rather than theoretical algorithmic complexity. Specifically, the assignment stage suffers from a severe IO bottleneck due to the massive explicit materialization of the N times K distance matrix in High Bandwidth Memory (HBM). Simultaneously, the centroid update stage is heavily penalized by hardware-level atomic write contention caused by irregular, scatter-style token aggregations. To bridge this performance gap, we propose flash-kmeans, an IO-aware and contention-free k-means implementation for modern GPU workloads. Flash-kmeans introduces two core kernel-level innovations: (1) FlashAssign, which fuses distance computation with an online argmin to completely bypass intermediate memory materialization; (2) sort-inverse update, which explicitly constructs an inverse mapping to transform high-contention atomic scatters into high-bandwidth, segment-level localized reductions. Furthermore, we integrate algorithm-system co-designs, including chunked-stream overlap and cache-aware compile heuristics, to ensure practical deployability. Extensive evaluations on NVIDIA H200 GPUs demonstrate that flash-kmeans achieves up to 17.9times end-to-end speedup over best baselines, while outperforming industry-standard libraries like cuML and FAISS by 33times and over 200times, respectively.

Large language models show promise for automated CUDA programming, however even the strongest coding models (e.g., Claude-Opus-4.6) may still fall short of expert-level, architecture-aware optimization. We introduce CUDAHercules, a benchmark that evaluates generated CUDA against end-to-end human-expert SOTA systems. It spans single kernels, module-level operators, full applications, and unsolved challenge tasks across Ampere, Hopper, and Blackwell GPUs, with end-to-end tasks gated by domain-specific semantic validators. Evaluating models such as Claude-Opus-4.6 and GPT-5.4 shows a large gap between runnable CUDA and expert CUDA engineering: models often compile and pass tests, but rarely recover the optimization strategies needed to match expert performance. Application semantics further reduce success, and iterative or tool-augmented feedback can improve correctness while drifting toward slow fallback implementations. These results show that automated CUDA programming remains far from fully solved and requires stronger hardware reasoning, better tool use, and training objectives that connect code understanding to hardware architecture-grounded intelligence.

Learned optimizers have been an active research topic over the past decade, with increasing progress toward practical, general-purpose optimizers that can serve as drop-in replacements for widely used methods like Adam. However, recent advances -- such as VeLO, which was meta-trained for 4000 TPU-months -- remain largely inaccessible to the broader community, in part due to their reliance on JAX and the absence of user-friendly packages for applying the optimizers after meta-training. To address this gap, we introduce PyLO, a PyTorch-based library that brings learned optimizers to the broader machine learning community through familiar, widely adopted workflows. Unlike prior work focused on synthetic or convex tasks, our emphasis is on applying learned optimization to real-world large-scale pre-training tasks. Our release includes a CUDA-accelerated version of the small_fc_lopt learned optimizer architecture from (Metz et al., 2022a), delivering substantial speedups -- from 39.36 to 205.59 samples/sec throughput for training ViT B/16 with batch size 32. PyLO also allows us to easily combine learned optimizers with existing optimization tools such as learning rate schedules and weight decay. When doing so, we find that learned optimizers can substantially benefit. Our code is available at https://github.com/Belilovsky-Lab/pylo

The rapid growth of large language model (LLM) inference services has increased the demand for efficient multi-tenant GPU scheduling. While modern inference runtimes such as vLLM improve throughput through continuous batching and optimized memory management, accurately estimating the runtime cost of heterogeneous inference requests remains challenging. In practice, admission-time workload estimates may deviate from observed execution behavior, leading to workload misclassification, queue imbalance, increased tail latency, and degraded Quality-of-Service (QoS). This paper presents DriftSched, a QoS-aware scheduling framework for multi-tenant LLM inference serving on NVIDIA L4 GPUs. DriftSched combines workload classification, token-budget estimation, tenant-aware queue management, and an online feedback mechanism to refine workload estimates using runtime observations. The framework evaluates FIFO, Priority, Weighted, Shortest-Job-First (SJF), and Aging Priority scheduling policies under heterogeneous multi-tenant workloads. Experimental results show that adaptive calibration reduces workload estimation error by an average of 38.8% (MAE) and 40.5% (RMSE), improving workload classification stability. Among all evaluated schedulers, SJF achieves the best overall performance, reducing median end-to-end latency by approximately 42% and P99 latency by approximately 16% relative to FIFO under sustained GPU contention. The results further indicate that scheduler selection has a greater impact on latency behavior than runtime calibration alone, while accurate workload characterization largely eliminates systematic estimation drift. This work contributes a reproducible framework for studying workload-estimation fidelity and QoS-aware scheduling in multi-tenant GPU inference systems.

This paper presents a comprehensive comparison of three dominant parallel programming models in High Performance Computing (HPC): Message Passing Interface (MPI), Open Multi-Processing (OpenMP), and Compute Unified Device Architecture (CUDA). Selecting optimal programming approaches for modern heterogeneous HPC architectures has become increasingly critical. We systematically analyze these models across multiple dimensions: architectural foundations, performance characteristics, domain-specific suitability, programming complexity, and recent advancements. We examine each model's strengths, weaknesses, and optimization techniques. Our investigation demonstrates that MPI excels in distributed memory environments with near-linear scalability for communication-intensive applications, but faces communication overhead challenges. OpenMP provides strong performance and usability in shared-memory systems and loop-centric tasks, though it is limited by shared memory contention. CUDA offers substantial performance gains for data-parallel GPU workloads, but is restricted to NVIDIA GPUs and requires specialized expertise. Performance evaluations across scientific simulations, machine learning, and data analytics reveal that hybrid approaches combining two or more models often yield optimal results in heterogeneous environments. The paper also discusses implementation challenges, optimization best practices, and emerging trends such as performance portability frameworks, task-based programming, and the convergence of HPC and Big Data. This research helps developers and researchers make informed decisions when selecting programming models for modern HPC applications, emphasizing that the best choice depends on application requirements, hardware, and development constraints.

As deep learning models scale, their training cost has surged significantly. Due to both hardware advancements and limitations in current software stacks, the need for data efficiency has risen. Data efficiency refers to the effective hiding of data access latency and the avoidance of unnecessary data movements. Major challenges arise from the growing disparity between GPU memory bandwidth and computational throughput, imminent GPU memory capacity limitations, and inefficiencies in the PyTorch software stack, including a lack of device-specific PCIe transfer optimizations and high-level domain-specific abstractions. To effectively mitigate these data inefficiencies for deep learning training, this dissertation analyzes data inefficiency in representative deep training tasks, specifically in graph neural networks (GNNs) and large language models (LLMs). It then proposes novel runtime and code generation techniques to mitigate these challenges and implements these optimizations seamlessly within the PyTorch stack while maintaining strong programmability and interoperability. First, PyTorch-Direct is devised to incorporate the GPU-centric PCIe data transfer paradigm in PyTorch for GNN training. Next, Hector intermediate representation (IR) and its code generator are proposed to introduce domain-specific high-level abstraction and systematically address memory-intensive performance challenges for relational GNNs. Finally, in LLM training, the throughput has been increasingly constrained by GPU memory capacity. To mitigate this, the SSDTrain offloading framework is designed and implemented. Together, these contributions show that code generation and runtime techniques can systematically mitigate the data management bottlenecks in deep learning training, which stem from the data-intensive nature of workloads and the oversimplification inherent in the deep learning training software stack.

Writing GPU kernels is a challenging task and critical for AI systems' efficiency. It is also highly iterative: domain experts write code and improve performance through execution feedback. Moreover, it presents verifiable rewards like correctness and speedup, making it a natural environment to apply Reinforcement Learning (RL). To explicitly incorporate the iterative nature of this process into training, we develop a flexible multi-turn RL recipe that addresses unique challenges encountered in real-world settings, such as learning from long trajectories and effective reward attribution across turns. We present Kevin - K(ernel D)evin, the first model trained with multi-turn RL for CUDA kernel generation and optimization. In our evaluation setup, Kevin shows significant gains over its base model (QwQ-32B), improving correctness of generated kernels (in pure CUDA) from 56% to 82% and mean speedup from 0.53x to 1.10x of baseline (PyTorch Eager), and surpassing frontier models like o4-mini (0.78x). Finally, we study its behavior across test-time scaling axes: we found scaling serial refinement more beneficial than parallel sampling. In particular, when given more refinement turns, Kevin shows a higher rate of improvement.

AI agents execute tens to hundreds of chained LLM calls per task, yet GPU schedulers treat each call as independent, discarding gigabytes of intermediate state between steps and inflating end-to-end latency by 3-8x. We argue that this request-level abstraction is fundamentally mismatched to compound AI workloads, and propose a shift to program-level scheduling: treating the entire agent workflow (not individual inference calls) as the first-class schedulable unit. We present SAGA, a distributed scheduler that implements this abstraction through three mechanisms: (1) Agent Execution Graphs that capture workflow structure to predict KV cache reuse across tool-call boundaries, achieving within 1.31x of Bélády's optimal offline policy; (2) session-affinity batching with work stealing that co-locates correlated requests while maintaining global load balance; and (3) Agent Fair Share, a task-completion-time fairness metric with provable bounded-deviation guarantees. On a 64-GPU cluster serving SWE-bench coding agents and WebArena browser tasks, SAGA reduces task completion time by 1.64x (geometric mean, p < 0.001) over vLLM v0.15.1 with prefix caching and affinity routing, while improving GPU memory utilization by 1.22x and achieving 99.2% SLO attainment under multi-tenant interference. These latency gains come at a quantified cost: approximately 30% lower peak throughput than throughput-optimal batch scheduling, a tradeoff appropriate for the latency-sensitive interactive deployments that dominate compound AI usage. Our results demonstrate that workflow-aware scheduling is essential for efficient compound AI serving.

In recent years, Large Language Models (LLMs) have exhibited remarkable capabilities, driving advancements in real-world applications. However, training LLMs on increasingly long input sequences imposes significant challenges due to high GPU memory and computational demands. Existing solutions face two key limitations: (1) memory reduction techniques, such as activation recomputation and CPU offloading, compromise training efficiency; (2) distributed parallelism strategies require excessive GPU resources, limiting the scalability of input sequence length. To address these gaps, we propose Adaptive Sequence Pipeline Parallel Offloading (SPPO), a novel LLM training framework that optimizes memory and computational resource efficiency for long-sequence training. SPPO introduces adaptive offloading, leveraging sequence-aware offloading, and two-level activation management to reduce GPU memory consumption without degrading the training efficiency. Additionally, SPPO develops an adaptive pipeline scheduling approach with a heuristic solver and multiplexed sequence partitioning to improve computational resource efficiency. Experimental results demonstrate that SPPO achieves up to 3.38x throughput improvement over Megatron-LM and DeepSpeed, realizing efficient training of a 7B LLM with sequence lengths of up to 4M tokens on only 128 A100 GPUs.

DistServe improves the performance of large language models (LLMs) serving by disaggregating the prefill and decoding computation. Existing LLM serving systems colocate the two phases and batch the computation of prefill and decoding across all users and requests. We find that this strategy not only leads to strong prefill-decoding interferences but also couples the resource allocation and parallelism plans for both phases. LLM applications often emphasize individual latency for each phase: time to first token (TTFT) for the prefill phase and time per output token (TPOT) of each request for the decoding phase. In the presence of stringent latency requirements, existing systems have to prioritize one latency over the other, or over-provision compute resources to meet both. DistServe assigns prefill and decoding computation to different GPUs, hence eliminating prefill-decoding interferences. Given the application's TTFT and TPOT requirements, DistServe co-optimizes the resource allocation and parallelism strategy tailored for each phase. DistServe also places the two phases according to the serving cluster's bandwidth to minimize the communication caused by disaggregation. As a result, DistServe significantly improves LLM serving performance in terms of the maximum rate that can be served within both TTFT and TPOT constraints on each GPU. Our evaluations show that on various popular LLMs, applications, and latency requirements, DistServe can serve 4.48x more requests or 10.2x tighter SLO, compared to state-of-the-art systems, while staying within latency constraints for > 90% of requests.

Large Language Models have revolutionized natural language processing, yet serving them efficiently in data centers remains challenging due to mixed workloads comprising latency-sensitive (LS) and best-effort (BE) jobs. Existing inference systems employ iteration-level first-come-first-served scheduling, causing head-of-line blocking when BE jobs delay LS jobs. We introduce QLLM, a novel inference system designed for Mixture of Experts (MoE) models, featuring a fine-grained, priority-aware preemptive scheduler. QLLM enables expert-level preemption, deferring BE job execution while minimizing LS time-to-first-token (TTFT). Our approach removes iteration-level scheduling constraints, enabling the scheduler to preempt jobs at any layer based on priority. Evaluations on an Nvidia A100 GPU show that QLLM significantly improves performance. It reduces LS TTFT by an average of 65.5times and meets the SLO at up to 7 requests/sec, whereas the baseline fails to do so under the tested workload. Additionally, it cuts LS turnaround time by up to 12.8times without impacting throughput. QLLM is modular, extensible, and seamlessly integrates with Hugging Face MoE models.

LLM serving relies on prefix caching to improve inference performance. As growing contexts push key-value (KV) cache footprint far beyond GPU HBM and CPU DRAM capacity, KV cache is increasingly offloaded to NVMe SSDs. Unfortunately, restoring KV cache from SSDs suffers from poor I/O performance and incurs significant GPU stalls. This is primarily because the fragmented GPU memory layout results in a massive number of tiny random I/Os, rendering the low-parallelism CPU a severe bottleneck even with GPU Direct Storage (GDS), which still relies on CPU intervention to initiate each I/O and thus remains CPU-centric. This paper presents Tutti, an efficient SSD-backed KV caching solution that eliminates CPU intervention from the critical data and I/O control paths between HBM and SSDs. At the core of Tutti is a GPU-centric KV cache object store, in which the CPU is only responsible for asynchronously loading I/O kernels once per layer to the GPU. Tutti saturates NVMe SSD bandwidth and reduces GPU stalls to near zero through the following designs: (i) we provide a GPU-native object abstraction that enables bulk KV cache transfers and management; (ii) we re-architect the GPU storage stack by introducing GPU io_uring to support asynchronous GPU direct object I/O; and (iii) we propose slack-aware I/O scheduling to avoid GPU resource contention. We have implemented Tutti and integrated it to vLLM. Extensive evaluation shows that compared to the state-of-the-art GDS-enabled, SSD-backed LMCache, Tutti reduces TTFT by 78.3% under strict SLO constraints and improves the achievable request rate by 2x. The serving cost is reduced by 27%. Tutti achieves nearly the same inference performance as DRAM-backed LMCache, while providing almost infinite capacity.

Optimizing complex systems, ranging from LLM prompts to multi-turn agents, traditionally requires labor-intensive manual iteration. We formalize this challenge as a stochastic generative optimization problem where a generative language model acts as the optimizer, guided by numerical rewards and text feedback to discover the best system. We introduce Prioritized Optimization with Local Contextual Aggregation (POLCA), a scalable framework designed to handle stochasticity in optimization -- such as noisy feedback, sampling minibatches, and stochastic system behaviors -- while effectively managing the unconstrained expansion of solution space. POLCA maintains a priority queue to manage the exploration-exploitation tradeoff, systematically tracking candidate solutions and their evaluation histories. To enhance efficiency, we integrate an varepsilon-Net mechanism to maintain parameter diversity and an LLM Summarizer to perform meta-learning across historical trials. We theoretically prove that POLCA converges to near-optimal candidate solutions under stochasticity. We evaluate our framework on diverse benchmarks, including τ-bench, HotpotQA (agent optimization), VeriBench (code translation) and KernelBench (CUDA kernel generation). Experimental results demonstrate that POLCA achieves robust, sample and time-efficient performance, consistently outperforming state-of-the-art algorithms in both deterministic and stochastic problems. The codebase for this work is publicly available at https://github.com/rlx-lab/POLCA.

Collocating deep learning training tasks improves GPU utilization but causes drastic slowdowns due to resource contention and risks Out-of-Memory (OOM) failures. Accurate memory estimation is essential for robust collocation, while GPU utilization -- a key proxy for resource contention -- enables interference-aware scheduling to reduce slowdowns and improve throughput. Existing GPU memory estimators span three paradigms -- analytical models, CPU-side libraries, and ML-based estimators -- each with distinct limitations: dependence on detailed model specifications, intrusive integration, poor generalization, and varying latency overhead. GPU heterogeneity further complicates estimation, as identical tasks can exhibit markedly different memory footprints across hardware generations. GPU utilization remains comparatively understudied, further complicated by the non-additive nature of utilization metrics and hardware sensitivity. We conduct a systematic analysis of representative estimators from each paradigm -- Horus, PyTorch FakeTensor, and our lightweight ML-based estimator -- evaluating accuracy, generalizability, and practical overhead. We construct a synthetic dataset spanning MLPs, CNNs, and Transformers with controlled architectural variations, and train MLP- and Transformer-based estimators for memory prediction. We further experiment with utilization estimation on the same dataset. Our evaluation reveals key tradeoffs and validates estimators against real-world unseen models. Significant challenges remain: analytical models are hardware-dependent, CPU-side libraries impose intrusive integration costs, and ML-based estimators struggle with cross-architecture generalization. We release all datasets, tools, and artifacts to support further research.

As Large Language Models (LLMs) are rapidly growing in popularity, LLM inference services must be able to serve requests from thousands of users while satisfying performance requirements. The performance of an LLM inference service is largely determined by the hardware onto which it is deployed, but understanding of which hardware will deliver on performance requirements remains challenging. In this work we present LLM-Pilot - a first-of-its-kind system for characterizing and predicting performance of LLM inference services. LLM-Pilot performs benchmarking of LLM inference services, under a realistic workload, across a variety of GPUs, and optimizes the service configuration for each considered GPU to maximize performance. Finally, using this characterization data, LLM-Pilot learns a predictive model, which can be used to recommend the most cost-effective hardware for a previously unseen LLM. Compared to existing methods, LLM-Pilot can deliver on performance requirements 33% more frequently, whilst reducing costs by 60% on average.

Transformers and large language models~(LLMs) have seen rapid adoption in all domains. Their sizes have exploded to hundreds of billions of parameters and keep increasing. Under these circumstances, the training of transformers is very expensive and often hits a ``memory wall'', i.e., even when using 3D parallelism (pipeline, tensor, data) and aggregating the memory of many GPUs, it is still not enough to hold the necessary data structures (model parameters, optimizer state, gradients, activations) in GPU memory. To compensate, state-of-the-art approaches offload the optimizer state, at least partially, to the host memory and perform hybrid CPU-GPU computations. However, the management of the combined host-GPU memory is often suboptimal and results in poor overlapping between data movements and computations. This leads to missed opportunities to simultaneously leverage the interconnect bandwidth and computational capabilities of CPUs and GPUs. In this paper, we leverage a key observation that the interleaving of the forward, backward and update phases generate fluctuations in the GPU memory utilization, which can be exploited to dynamically move a part of the optimizer state between the host and the GPU memory at each iteration. To this end, we design and implement \proj, a novel technique to split the LLM into subgroups, whose update phase is scheduled on either the CPU or the GPU based on our proposed performance model that addresses the trade-off between data movement cost, acceleration on the GPUs vs the CPUs, and competition for shared resources. We integrate our approach with DeepSpeed and demonstrate 2.5times faster iterations over state-of-the-art approaches using extensive experiments.

Large Language Models (LLMs) for Generative AI have achieved remarkable progress, evolving into sophisticated and versatile tools widely adopted across various domains and applications. However, the substantial memory overhead caused by their vast number of parameters, combined with the high computational demands of the attention mechanism, poses significant challenges in achieving low latency and high throughput for LLM inference services. Recent advancements, driven by groundbreaking research, have significantly accelerated progress in this field. This paper provides a comprehensive survey of these methods, covering fundamental instance-level approaches, in-depth cluster-level strategies, emerging scenario directions, and other miscellaneous but important areas. At the instance level, we review model placement, request scheduling, decoding length prediction, storage management, and the disaggregation paradigm. At the cluster level, we explore GPU cluster deployment, multi-instance load balancing, and cloud service solutions. For emerging scenarios, we organize the discussion around specific tasks, modules, and auxiliary methods. To ensure a holistic overview, we also highlight several niche yet critical areas. Finally, we outline potential research directions to further advance the field of LLM inference serving.

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

Multimodal large language models (MLLMs) extend LLMs with visual understanding through a three-stage pipeline: multimodal preprocessing, vision encoding, and LLM inference. While these stages enhance capability, they introduce significant system bottlenecks. First, multimodal preprocessing-especially video decoding-often dominates Time-to-First-Token (TTFT). Most systems rely on CPU-based decoding, which severely limits throughput, while existing GPU-based approaches prioritize throughput-oriented parallelism and fail to meet the latency-sensitive requirements of MLLM inference. Second, the vision encoder is a standalone, compute-intensive stage that produces visual embeddings and cannot be co-batched with LLM prefill or decoding. This heterogeneity forces inter-stage blocking and increases token-generation latency. Even when deployed on separate GPUs, these stages underutilize available compute and memory resources, reducing overall utilization and constraining system throughput. To address these challenges, we present FlashCodec and UnifiedServe, two complementary designs that jointly optimize the end-to-end MLLM pipeline. FlashCodec accelerates the multimodal preprocessing stage through collaborative multi-GPU video decoding, reducing decoding latency while preserving high throughput. UnifiedServe optimizes the vision-to-text and inference stages using a logically decoupled their execution to eliminate inter-stage blocking, yet physically sharing GPU resources to maximize GPU system utilization. By carefully orchestrating execution across stages and minimizing interference, UnifiedServe Together, our proposed framework forms an end-to-end optimized stack that can serve up to 3.0times more requests or enforce 1.5times tighter SLOs, while achieving up to 4.4times higher throughput compared to state-of-the-art systems.

Low-rank adaptation (LoRA) has become an important and popular method to adapt pre-trained models to specific domains. We present Punica, a system to serve multiple LoRA models in a shared GPU cluster. Punica contains a new CUDA kernel design that allows batching of GPU operations for different LoRA models. This allows a GPU to hold only a single copy of the underlying pre-trained model when serving multiple, different LoRA models, significantly enhancing GPU efficiency in terms of both memory and computation. Our scheduler consolidates multi-tenant LoRA serving workloads in a shared GPU cluster. With a fixed-sized GPU cluster, our evaluations show that Punica achieves 12x higher throughput in serving multiple LoRA models compared to state-of-the-art LLM serving systems while only adding 2ms latency per token. Punica is open source at https://github.com/punica-ai/punica .

Reinforcement Learning (RL) has become the most effective post-training approach for improving the capabilities of Large Language Models (LLMs). In practice, because of the high demands on latency and memory, it is particularly challenging to develop an efficient RL framework that reliably manages policy models with hundreds to thousands of billions of parameters. In this paper, we present LlamaRL, a fully distributed, asynchronous RL framework optimized for efficient training of large-scale LLMs with various model sizes (8B, 70B, and 405B parameters) on GPU clusters ranging from a handful to thousands of devices. LlamaRL introduces a streamlined, single-controller architecture built entirely on native PyTorch, enabling modularity, ease of use, and seamless scalability to thousands of GPUs. We also provide a theoretical analysis of LlamaRL's efficiency, including a formal proof that its asynchronous design leads to strict RL speed-up. Empirically during the Llama 3 post-training, by leveraging best practices such as colocated model offloading, asynchronous off-policy training, and distributed direct memory access for weight synchronization, LlamaRL achieves significant efficiency gains -- up to 10.7x speed-up compared to DeepSpeed-Chat-like systems on a 405B-parameter policy model. Furthermore, the efficiency advantage continues to grow with increasing model scale, demonstrating the framework's suitability for future large-scale RL training.

Scaling Mixture-of-Experts (MoE) training introduces systems challenges absent in dense models. Because each token activates only a subset of experts, this sparsity allows total parameters to grow much faster than per-token computation, creating coupled constraints across memory, communication, and computation. Optimizing one dimension often shifts pressure to another, demanding co-design across the full system stack. We address these challenges for MoE training through integrated optimizations spanning memory (fine-grained recomputation, offloading, etc.), communication (optimized dispatchers, overlapping, etc.), and computation (Grouped GEMM, fusions, CUDA Graphs, etc.). The framework also provides Parallel Folding for flexible multi-dimensional parallelism, low-precision training support for FP8 and NVFP4, and efficient long-context training. On NVIDIA GB300 and GB200, it achieves 1,233/1,048 TFLOPS/GPU for DeepSeek-V3-685B and 974/919 TFLOPS/GPU for Qwen3-235B. As a performant, scalable, and production-ready open-source solution, it has been used across academia and industry for training MoE models ranging from billions to trillions of parameters on clusters scaling up to thousands of GPUs. This report explains how these techniques work, their trade-offs, and their interactions at the systems level, providing practical guidance for scaling MoE models with Megatron Core.

We explore reinforcement learning (RL) techniques to enhance reasoning within targeted problem spaces in large language models (LLMs) under memory and compute constraints. Our focus is on critic-free methods compatible with LoRA fine-tuning on a single 40GB GPU, a common limitation in academic settings. We introduce S-GRPO, a memory-efficient variant of Group Relative Policy Optimization, and T-SPMO, a token-level prefix matching strategy for fine-grained credit assignment. Despite limited resources, when used to fine-tune Qwen2-1.5B both methods significantly improve SVAMP benchmark accuracy from 46% to above 70% using LoRA training. T-SPMO also excels in multi-digit multiplication tasks, underscoring the potential of RL fine-tuning under hardware constraints. Additionally, we find that our full-token GRPO baseline under LoRA fine-tuning did not improve model performance (compared to base model) on either task, suggesting that our memory-efficient methods may act as a form of regularization that stabilizes training when only a small subset of parameters are updated.

Large language model (LLM) training and finetuning are often bottlenecked by limited GPU memory. While existing projection-based optimization methods address this by projecting gradients into a lower-dimensional subspace to reduce optimizer state memory, they typically rely on dense projection matrices, which can introduce computational and memory overheads. In this work, we propose Grass (GRAdient Stuctured Sparsification), a novel approach that leverages sparse projections to transform gradients into structured sparse updates. This design not only significantly reduces memory usage for optimizer states but also minimizes gradient memory footprint, computation, and communication costs, leading to substantial throughput improvements. Extensive experiments on pretraining and finetuning tasks demonstrate that Grass achieves competitive performance to full-rank training and existing projection-based methods. Notably, Grass enables half-precision pretraining of a 13B parameter LLaMA model on a single 40GB A100 GPU--a feat infeasible for previous methods--and yields up to a 2times throughput improvement on an 8-GPU system. Code can be found at https://github.com/aashiqmuhamed/GRASS .

Serverless computing has grown rapidly for serving Large Language Model (LLM) inference due to its pay-as-you-go pricing, fine-grained GPU usage, and rapid scaling. However, our analysis reveals that current serverless can effectively serve general LLM but fail with Low-Rank Adaptation (LoRA) inference due to three key limitations: 1) massive parameter redundancy among functions where 99% of weights are unnecessarily duplicated, 2) costly artifact loading latency beyond LLM loading, and 3) magnified resource contention when serving multiple LoRA LLMs. These inefficiencies lead to massive GPU wastage, increased Time-To-First-Token (TTFT), and high monetary costs. We propose ServerlessLoRA, a novel serverless inference system designed for faster and cheaper LoRA LLM serving. ServerlessLoRA enables secure backbone LLM sharing across isolated LoRA functions to reduce redundancy. We design a pre-loading method that pre-loads comprehensive LoRA artifacts to minimize cold-start latency. Furthermore, ServerlessLoRA employs contention aware batching and offloading to mitigate GPU resource conflicts during bursty workloads. Experiment on industrial workloads demonstrates that ServerlessLoRA reduces TTFT by up to 86% and cuts monetary costs by up to 89% compared to state-of-the-art LLM inference solutions.

Matrix-based preconditioned optimizers, such as Muon, have recently been shown to be more efficient than scalar-based optimizers for training large-scale neural networks, including large language models (LLMs). On the other hand, recent benchmarks on optimizers for LLM pre-training have demonstrated that variance-reduction techniques such as MARS can achieve substantial speedups over standard optimizers that do not employ variance reduction. In this paper, to achieve the best of both worlds, we introduce MARS-M, a new optimizer that integrates the variance reduction technique in MARS with Muon. Under standard regularity conditions, we prove that Muon-M converges to a first-order stationary point at a rate of mathcal{O}(T^{-1/3}), which improves upon mathcal{O}(T^{-1/4}) rate attained by Muon. Our empirical results on language modeling and computer vision tasks demonstrate that MARS-M consistently yields lower losses and improved performance across various downstream benchmarks. The implementation of MARS-M is available at https://github.com/AGI-Arena/MARS/MARS_M.

Large Language Models (LLMs) have presented impressive performance across several transformative tasks. However, it is non-trivial to efficiently utilize large-scale cluster resources to develop LLMs, often riddled with numerous challenges such as frequent hardware failures, intricate parallelization strategies, and imbalanced resource utilization. In this paper, we present an in-depth characterization study of a six-month LLM development workload trace collected from our GPU datacenter Acme. Specifically, we investigate discrepancies between LLMs and prior task-specific Deep Learning (DL) workloads, explore resource utilization patterns, and identify the impact of various job failures. Our analysis summarizes hurdles we encountered and uncovers potential opportunities to optimize systems tailored for LLMs. Furthermore, we introduce our system efforts: (1) fault-tolerant pretraining, which enhances fault tolerance through LLM-involved failure diagnosis and automatic recovery. (2) decoupled scheduling for evaluation, which achieves timely performance feedback via trial decomposition and scheduling optimization.

Large Language Models (LLMs) have showcased exceptional performance across a wide array of Natural Language Processing (NLP) tasks. Fine-tuning techniques are commonly utilized to tailor pre-trained models to specific applications. While methods like LoRA have effectively tackled GPU memory constraints during fine-tuning, their applicability is often restricted to limited performance, especially on multi-task. On the other hand, Mix-of-Expert (MoE) models, such as Mixtral 8x7B, demonstrate remarkable performance across multiple NLP tasks while maintaining a reduced parameter count. However, the resource requirements of these MoEs still challenging, particularly for consumer-grade GPUs only have limited VRAM. To address these challenge, we propose MixLoRA, an innovative approach aimed at constructing a resource-efficient sparse MoE model based on LoRA. MixLoRA inserts multiple LoRA-based experts within the feed-forward network block of a frozen pre-trained dense model through fine-tuning, employing a commonly used top-k router. Unlike other LoRA based MoE methods, MixLoRA enhances model performance by utilizing independently configurable attention-layer LoRA adapters, supporting the use of LoRA and its variants for the construction of experts, and applying auxiliary load balance loss to address the imbalance problem of the router. In experiments, MixLoRA achieves commendable performance across all evaluation metrics in both single-task and multi-task learning scenarios. Implemented within the m-LoRA framework, MixLoRA enables parallel fine-tuning of multiple mixture-of-experts models on a single 24GB consumer-grade GPU without quantization, thereby reducing GPU memory consumption by 41\% and latency during the training process by 17\%.

The scaling of Large Language Models (LLMs) drives interest in matrix-based optimizers (e.g., Shampoo, Muon, SOAP) for their convergence efficiency; yet their requirement for holistic updates conflicts with the tensor fragmentation in distributed frameworks like Megatron. Existing solutions are suboptimal: synchronous approaches suffer from computational redundancy, while layer-wise partitioning fails to reconcile this conflict without violating the geometric constraints of efficient communication primitives. To bridge this gap, we propose Canzona, a Unified, Asynchronous, and Load-Balanced framework that decouples logical optimizer assignment from physical parameter distribution. For Data Parallelism, we introduce an alpha-Balanced Static Partitioning strategy that respects atomicity while neutralizing the load imbalance. For Tensor Parallelism, we design an Asynchronous Compute pipeline utilizing Micro-Group Scheduling to batch fragmented updates and hide reconstruction overhead. Extensive evaluations on the Qwen3 model family (up to 32B parameters) on 256 GPUs demonstrate that our approach preserves the efficiency of established parallel architectures, achieving a 1.57x speedup in end-to-end iteration time and reducing optimizer step latency by 5.8x compared to the baseline.

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

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.

Large language models (LLMs) are notoriously memory-intensive during training, particularly with the popular AdamW optimizer. This memory burden necessitates using more or higher-end GPUs or reducing batch sizes, limiting training scalability and throughput. To address this, various memory-efficient optimizers have been proposed to reduce optimizer memory usage. However, they face critical challenges: (i) reliance on costly SVD operations; (ii) significant performance trade-offs compared to AdamW; and (iii) still substantial optimizer memory overhead to maintain competitive performance. In this work, we identify that AdamW's learning rate adaptation rule can be effectively coarsened as a structured learning rate update. Based on this insight, we propose Approximated Gradient Scaling for Memory-Efficient LLM Optimization (APOLLO), which approximates learning rate scaling using an auxiliary low-rank optimizer state based on pure random projection. This structured learning rate update rule makes APOLLO highly tolerant to further memory reductions while delivering comparable pre-training performance. Even its rank-1 variant, APOLLO-Mini, achieves superior pre-training performance compared to AdamW with SGD-level memory costs. Extensive experiments demonstrate that the APOLLO series performs on-par with or better than AdamW, while achieving greater memory savings by nearly eliminating the optimization states of AdamW. These savings provide significant system-level benefits: (1) Enhanced Throughput: 3x throughput on an 8xA100-80GB setup compared to AdamW by supporting 4x larger batch sizes. (2) Improved Model Scalability: Pre-training LLaMA-13B with naive DDP on A100-80GB GPUs without system-level optimizations. (3) Low-End GPU Friendly Pre-training: Pre-training LLaMA-7B on a single GPU using less than 12 GB of memory with weight quantization.

Large Language Models (LLMs) offer state-of-the-art performance in natural language understanding and generation tasks. However, the deployment of leading commercial models for specialized tasks, such as e-commerce, is often hindered by high computational costs, latency, and operational expenses. This paper investigates the viability of smaller, open-weight models as a resource-efficient alternative. We present a methodology for optimizing a one-billion-parameter Llama 3.2 model for multilingual e-commerce intent recognition. The model was fine-tuned using Quantized Low-Rank Adaptation (QLoRA) on a synthetically generated dataset designed to mimic real-world user queries. Subsequently, we applied post-training quantization techniques, creating GPU-optimized (GPTQ) and CPU-optimized (GGUF) versions. Our results demonstrate that the specialized 1B model achieves 99% accuracy, matching the performance of the significantly larger GPT-4.1 model. A detailed performance analysis revealed critical, hardware-dependent trade-offs: while 4-bit GPTQ reduced VRAM usage by 41%, it paradoxically slowed inference by 82% on an older GPU architecture (NVIDIA T4) due to dequantization overhead. Conversely, GGUF formats on a CPU achieved a speedup of up to 18x in inference throughput and a reduction of over 90% in RAM consumption compared to the FP16 baseline. We conclude that small, properly optimized open-weight models are not just a viable but a more suitable alternative for domain-specific applications, offering state-of-the-art accuracy at a fraction of the computational cost.

Evolutionary multiobjective optimization (EMO) has made significant strides over the past two decades. However, as problem scales and complexities increase, traditional EMO algorithms face substantial performance limitations due to insufficient parallelism and scalability. While most work has focused on algorithm design to address these challenges, little attention has been given to hardware acceleration, thereby leaving a clear gap between EMO algorithms and advanced computing devices, such as GPUs. To bridge the gap, we propose to parallelize EMO algorithms on GPUs via the tensorization methodology. By employing tensorization, the data structures and operations of EMO algorithms are transformed into concise tensor representations, which seamlessly enables automatic utilization of GPU computing. We demonstrate the effectiveness of our approach by applying it to three representative EMO algorithms: NSGA-III, MOEA/D, and HypE. To comprehensively assess our methodology, we introduce a multiobjective robot control benchmark using a GPU-accelerated physics engine. Our experiments show that the tensorized EMO algorithms achieve speedups of up to 1113x compared to their CPU-based counterparts, while maintaining solution quality and effectively scaling population sizes to hundreds of thousands. Furthermore, the tensorized EMO algorithms efficiently tackle complex multiobjective robot control tasks, producing high-quality solutions with diverse behaviors. Source codes are available at https://github.com/EMI-Group/evomo.

Large language models (LLMs) require vast amounts of GPU compute to train, but limited availability and high costs of GPUs make homogeneous clusters impractical for many organizations. Instead, assembling heterogeneous clusters by pooling together GPUs of different generations allows them to achieve higher aggregate compute and make use of all available GPUs. However, training on heterogeneous clusters presents several challenges, including load balancing across GPUs, optimizing memory usage to accommodate varying memory capacities, and ensuring communication-efficient training over diverse network interconnects potentially spanning multiple datacenters. In this paper, we make the case that efficient training on heterogeneous clusters requires (1) the integration of pipeline parallelism and data parallelism in a manner that is both communication- and memory-efficient, and (2) a more adaptable configuration of pipeline and data parallelism, which includes the capability to flexibly partition GPUs into asymmetric pipeline parallel stages and to incorporate heterogeneous GPUs within the same data parallelism group. We propose Zorse, the first system to unify all these capabilities while incorporating a planner that automatically configures training strategies for a given workload. Our evaluation shows that Zorse significantly outperforms state-of-the-art systems in heterogeneous training scenarios.

We introduce Mirage Persistent Kernel (MPK), the first compiler and runtime system that automatically transforms multi-GPU model inference into a single high-performance megakernel. MPK introduces an SM-level graph representation that captures data dependencies at the granularity of individual streaming multiprocessors (SMs), enabling cross-operator software pipelining, fine-grained kernel overlap, and other previously infeasible GPU optimizations. The MPK compiler lowers tensor programs into highly optimized SM-level task graphs and generates optimized CUDA implementations for all tasks, while the MPK in-kernel parallel runtime executes these tasks within a single mega-kernel using decentralized scheduling across SMs. Together, these components provide end-to-end kernel fusion with minimal developer effort, while preserving the flexibility of existing programming models. Our evaluation shows that MPK significantly outperforms existing kernel-per-operator LLM serving systems by reducing end-to-end inference latency by up to 1.7x, pushing LLM inference performance close to hardware limits. MPK is publicly available at https://github.com/mirage-project/mirage.

Convolution is the most time-consuming operation in deep neural network operations, so its performance is critical to the overall performance of the neural network. The commonly used methods for convolution on GPU include the general matrix multiplication (GEMM)-based convolution and the direct convolution. GEMM-based convolution relies on the im2col algorithm, which results in a large memory footprint and reduced performance. Direct convolution does not have the large memory footprint problem, but the performance is not on par with GEMM-based approach because of the discontinuous memory access. This paper proposes a window-order-based convolution paradigm on GPU, called im2win, which not only reduces memory footprint but also offers continuous memory accesses, resulting in improved performance. Furthermore, we apply a range of optimization techniques on the convolution CUDA kernel, including shared memory, tiling, micro-kernel, double buffer, and prefetching. We compare our implementation with the direct convolution, and PyTorch's GEMM-based convolution with cuBLAS and six cuDNN-based convolution implementations, with twelve state-of-the-art DNN benchmarks. The experimental results show that our implementation 1) uses less memory footprint by 23.1% and achieves 3.5times TFLOPS compared with cuBLAS, 2) uses less memory footprint by 32.8% and achieves up to 1.8times TFLOPS compared with the best performant convolutions in cuDNN, and 3) achieves up to 155times TFLOPS compared with the direct convolution. We further perform an ablation study on the applied optimization techniques and find that the micro-kernel has the greatest positive impact on performance.

Translating complex reinforcement learning (RL) environments into high-performance implementations has traditionally required months of specialized engineering. We present a reusable recipe - a generic prompt template, hierarchical verification, and iterative agent-assisted repair - that produces semantically equivalent high-performance environments for <$10 in compute cost. We demonstrate three distinct workflows across five environments. Direct translation (no prior performance implementation exists): EmuRust (1.5x PPO speedup via Rust parallelism for a Game Boy emulator) and PokeJAX, the first GPU-parallel Pokemon battle simulator (500M SPS random action, 15.2M SPS PPO; 22,320x over the TypeScript reference). Translation verified against existing performance implementations: throughput parity with MJX (1.04x) and 5x over Brax at matched GPU batch sizes (HalfCheetah JAX); 42x PPO (Puffer Pong). New environment creation: TCGJax, the first deployable JAX Pokemon TCG engine (717K SPS random action, 153K SPS PPO; 6.6x over the Python reference), synthesized from a web-extracted specification. At 200M parameters, the environment overhead drops below 4% of training time. Hierarchical verification (property, interaction, and rollout tests) confirms semantic equivalence for all five environments; cross-backend policy transfer confirms zero sim-to-sim gap for all five environments. TCGJax, synthesized from a private reference absent from public repositories, serves as a contamination control for agent pretraining data concerns. The paper contains sufficient detail - including representative prompts, verification methodology, and complete results - that a coding agent could reproduce the translations directly from the manuscript.