Reinforcement learning (RL) post-training for Large Language Models (LLMs) is now scaling to large clusters and running for extended durations to enhance model reasoning performance. However, the scalability of existing RL frameworks is limited, as extreme long-tail skewness in RL trajectory generation causes severe GPU underutilization. Current asynchronous RL systems attempt to mitigate this, but they rely on global weight synchronization between the actor and all rollouts, which creates a rigid model update schedule. This global synchronization is ill-suited for the highly skewed and evolving distribution of trajectory generation latency in RL training, crippling training efficiency. Our key insight is that efficient scaling requires breaking this lockstep through trajectory-level asynchrony, which generates and consumes each trajectory independently. We propose Laminar, a scalable and robust RL post-training system built on a fully decoupled architecture. First, we replace global updates with a tier of relay workers acting as a distributed parameter service. This enables asynchronous and fine-grained weight synchronization, allowing rollouts to pull the latest weight anytime without stalling the actor's training loop. Second, a dynamic repack mechanism consolidates long-tail trajectories onto a few dedicated rollouts, maximizing generation throughput. The fully decoupled design also isolates failures, ensuring robustness for long-running jobs. Our evaluation on a 1024-GPU cluster shows that Laminar achieves up to 5.48$\times$ training throughput speedup over state-of-the-art systems, while reducing model convergence time.

Modern frameworks for training large foundation models (LFMs) employ dataloaders in a data-parallel manner, with each loader processing a disjoint subset of training data. When preparing data for LFM training that originates from multiple, distinct sources, two fundamental challenges arise. First, due to the quadratic computational complexity of the attention operator, the non-uniform sample distribution over data-parallel ranks leads to significant workload imbalance among dataloaders, degrading the training efficiency. Second, supporting diverse data sources requires per-dataset file access states that are redundantly replicated across parallel loaders, consuming excessive memory. This also hinders dynamic data mixing (e.g., curriculum learning) and causes redundant access/memory overhead in hybrid parallelism.

Modern large language models (LLMs) serving systems address distributed deployment challenges through two key techniques: distributed model partitioning for parallel computation across accelerators and quantization for reducing parameter size. While existing systems assume homogeneous GPU environments, we reveal significant untapped potential in heterogeneous systems with mixed-capacity accelerators where two critical limitations persist: (1) uniform partitioning and quantization strategies fail to adapt to hardware heterogeneity, exacerbating resource imbalance, and (2) decoupled optimization of partitioning and quantization overlooks critical performance synergies between these techniques. We present SplitQuant, a phase-aware distributed serving system that co-optimizes mixedprecision quantization, phase-aware model partitioning, and micro-batch sizing for heterogeneous environments. Our approach combines analytical modeling of quality-runtime tradeoffs with a lightweight planning algorithm to maximize throughput while preserving user-specified model quality targets. Evaluations across 10 production clusters show SplitQuant achieves up to$2.34 \times(1.61 \times$mean) higher throughput than state-of-theart approaches without violating accuracy targets. Our results underscore the value of co-designing quantization and model partitioning strategies for heterogeneous environments.

The training scale of large language models (LLMs) has reached tens of thousands of GPUs and is still continuously expanding, enabling faster learning of larger models. Accompanying the expansion of the resource scale is the prevalence of failures (CUDA error, NaN values, job hang, etc.), which poses significant challenges to training stability. Any large-scale LLM training infrastructure should strive for minimal training interruption, efficient fault diagnosis, and effective failure tolerance to enable highly efficient continuous training. This paper presents ByteRobust, a large-scale GPU infrastructure management system tailored for robust and stable training of LLMs. It exploits the uniqueness of LLM training process and gives top priorities to detecting and recovering failures in a routine manner. Leveraging parallelisms and characteristics of LLM training, ByteRobust enables high-capacity fault tolerance, prompt fault demarcation, and localization with an effective data-driven approach, comprehensively ensuring continuous and efficient training of LLM tasks. ByteRobust is deployed on a production GPU platform with over 200,000 GPUs and advances the state of the art in training robustness by achieving 97% ETTR for a three-month training job on 9,600 GPUs.

A number of production deep learning clusters have attempted to explore inference hardware for DNN training, at the off-peak serving hours with many inference GPUs idling. Conducting DNN training with a combination of heterogeneous training and inference GPUs, known as hybrid device training, presents considerable challenges due to disparities in compute capability and significant differences in memory capacity. We propose QSync, a training system that enables efficient synchronous data-parallel DNN training over hybrid devices by strategically exploiting quantized operators. According to each device’s available resource capacity, QSync selects a quantization-minimized setting for operators in the distributed DNN training graph, minimizing model accuracy degradation but keeping the training efficiency brought by quantization. We carefully design a predictor with a bi-directional mixed-precision indicator to reflect the sensitivity of DNN layers on fixed-point and floating-point low-precision operators, a replayer with a neighborhood-aware cost mapper to accurately estimate the latency of distributed hybrid mixed-precision training, and then an allocator that efficiently synchronizes workers with minimized model accuracy degradation. QSync bridges the computational graph on PyTorch to an optimized backend for quantization kernel performance and flexible support for various GPU architectures. Extensive experiments show that QSync’s predictor can accurately simulate distributed mixed-precision training with < 5% error, with a consistent 0.27 − 1.03% accuracy improvement over the from-scratch training tasks compared to uniform precision.

The immense sizes of Large-scale language models (LLMs) have led to high resource demand and cost for running the models. Though the models are largely served using uniform high-caliber GPUs nowadays, utilizing a heterogeneous cluster with a mix of available high- and low-capacity GPUs can potentially substantially reduce the serving cost. This paper proposes LLM-PQ, a system that advocates adaptive model quantization and phase-aware partition to improve LLM serving efficiency on heterogeneous GPU clusters. Extensive experiments on production inference workloads demonstrate throughput improvement in inference, showing great advantages over state-of-the-art works.

Machine Learning (ML) models are progressively deployed in many real-world applications to perform a wide range of tasks, but are exposed to the security and privacy threats which aim to infer the details and even steal the functionality of the ML models. Despite extensive attacking efforts which rely on white-box or gray-box access, how to perform attacks with black-box access continues to be elusive. Aspiring to fill this gap, we move one step further and present ML-Stealer that can steal the functionality of any type of ML models with mere black-box access. With two algorithm designs, namely, synthetic data generation and replica model construction, ML-Stealer can construct a deep neural network (DNN)-based replica model which has the similar prediction functionality to the victim ML model. ML-Stealer does not require any knowledge about the victim model, nor does it enforce the access to statistical information or samples of the victim's training data. Experiment results demonstrate that ML-Stealer can achieve the consistent prediction results with the victim model of an averaged testing accuracy of 85.6%, and up to 93.6% at best.