Thousand-GPU Large-Scale Training and Optimization Recipe for AI-Native Cloud Embodied Intelligence Infrastructure

• URL: http://arxiv.org/abs/2603.11101v2
• Date: Wed, 18 Mar 2026 08:30:14 GMT
• Title: Thousand-GPU Large-Scale Training and Optimization Recipe for AI-Native Cloud Embodied Intelligence Infrastructure
• Abstract summary: Embodied intelligence is a key step towards Artificial General Intelligence (AGI)<n>We have launched a cloud-based, thousand-GPU distributed training platform for embodied intelligence.<n>We have built an end-to-end evaluation system, creating a closed loop from training to simulation to assessment.
• Score: 24.0829526239246
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Embodied intelligence is a key step towards Artificial General Intelligence (AGI), yet its development faces multiple challenges including data, frameworks, infrastructure, and evaluation systems. To address these issues, we have, for the first time in the industry, launched a cloud-based, thousand-GPU distributed training platform for embodied intelligence, built upon the widely adopted LeRobot framework, and have systematically overcome bottlenecks across the entire pipeline. At the data layer, we have restructured the data pipeline to optimize the flow of embodied training data. In terms of training, for the GR00T-N1.5 model, utilizing thousand-GPU clusters and data at the scale of hundreds of millions, the single-round training time has been reduced from 15 hours to just 22 minutes, achieving a 40-fold speedup. At the model layer, by combining variable-length FlashAttention and Data Packing, we have moved from sample redundancy to sequence integration, resulting in a 188% speed increase; π-0.5 attention optimization has accelerated training by 165%; and FP8 quantization has delivered a 140% speedup. On the infrastructure side, relying on high-performance storage, a 3.2T RDMA network, and a Ray-driven elastic AI data lake, we have achieved deep synergy among data, storage, communication, and computation. We have also built an end-to-end evaluation system, creating a closed loop from training to simulation to assessment. This framework has already been fully validated on thousand-GPU clusters, laying a crucial technical foundation for the development and application of next-generation autonomous intelligent robots, and is expected to accelerate the arrival of the era of human-machine integration.

Related papers

• SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD [91.71050339479261]
  This report presents an end-to-end optimization practice on the Ascend NPU SuperPOD.<n>We develop a hierarchical optimization framework spanning model-level parallelism, computation-communication orchestration, and low-level kernel execution.<n>The resulting system achieves 34.22% Model FLOPs Utilization (MFU) with a 2.93x improvement over the open-source baseline recipe.
  arXiv  Detail & Related papers  (2026-07-22T13:49:17Z)
• Breaking the Training Barrier of Billion-Parameter Universal Machine Learning Interatomic Potentials [25.963136125890298]
  We introduce MatRIS-MoE and Janus, a pioneering high-dimensional distributed training framework for uMLIPs with hardware-aware optimizations.<n>Our code attains a peak performance of 1.2/1.0 EFLOPS (24%/35.5% of theoretical peak) in single precision at over 90% parallel efficiency.<n>This work establishes a new high-water mark for AI-for-Science (AI4S) foundation models at Exascale.
  arXiv  Detail & Related papers  (2026-04-17T08:20:32Z)
• A Pragmatic VLA Foundation Model [66.76609538850478]
  We develop LingBot-VLA with around 20,000 hours of real-world data from 9 popular dual-arm robot configurations.<n>Our model achieves clear superiority over competitors, showcasing its strong performance and broad generalizability.<n>To advance the field of robot learning, we provide open access to the code, base model, and benchmark data.
  arXiv  Detail & Related papers  (2026-01-26T17:08:04Z)
• Data Scaling Laws for End-to-End Autonomous Driving [83.85463296830743]
  We evaluate the performance of a simple end-to-end driving architecture on internal driving datasets ranging in size from 16 to 8192 hours.<n>Specifically, we investigate how much additional training data is needed to achieve a target performance gain.
  arXiv  Detail & Related papers  (2025-04-06T03:23:48Z)
• FusionLLM: A Decentralized LLM Training System on Geo-distributed GPUs with Adaptive Compression [55.992528247880685]
  Decentralized training faces significant challenges regarding system design and efficiency. We present FusionLLM, a decentralized training system designed and implemented for training large deep neural networks (DNNs) We show that our system and method can achieve 1.45 - 9.39x speedup compared to baseline methods while ensuring convergence.
  arXiv  Detail & Related papers  (2024-10-16T16:13:19Z)
• High Throughput Training of Deep Surrogates from Large Ensemble Runs [38.018735015291156]
  Deep surrogates are generally trained in a supervised manner from limited amounts of data slowly generated by the same solver they intend to accelerate. We propose an open-source framework that enables the online training of these models from a large ensemble run of simulations. Experiment on training a fully connected network as a surrogate for the heat equation shows the proposed approach enables training on 8TB of data in 2 hours with an accuracy improved by 47% and a batch throughput multiplied by 13 compared to a traditional offline procedure.
  arXiv  Detail & Related papers  (2023-09-28T09:34:52Z)
• LearningGroup: A Real-Time Sparse Training on FPGA via Learnable Weight Grouping for Multi-Agent Reinforcement Learning [2.0625936401496237]
  Multi-agent reinforcement learning (MARL) is a powerful technology to construct interactive artificial intelligent systems. We present a real-time sparse training acceleration system named LearningGroup. Our system minimizes the cycle time and memory footprint for sparse data generation up to 5.72x and 6.81x.
  arXiv  Detail & Related papers  (2022-10-29T15:09:34Z)
• Accelerating Training and Inference of Graph Neural Networks with Fast Sampling and Pipelining [58.10436813430554]
  Mini-batch training of graph neural networks (GNNs) requires a lot of computation and data movement. We argue in favor of performing mini-batch training with neighborhood sampling in a distributed multi-GPU environment. We present a sequence of improvements to mitigate these bottlenecks, including a performance-engineered neighborhood sampler. We also conduct an empirical analysis that supports the use of sampling for inference, showing that test accuracies are not materially compromised.
  arXiv  Detail & Related papers  (2021-10-16T02:41:35Z)
• Large Batch Simulation for Deep Reinforcement Learning [101.01408262583378]
  We accelerate deep reinforcement learning-based training in visually complex 3D environments by two orders of magnitude over prior work. We realize end-to-end training speeds of over 19,000 frames of experience per second on a single and up to 72,000 frames per second on a single eight- GPU machine. By combining batch simulation and performance optimizations, we demonstrate that Point navigation agents can be trained in complex 3D environments on a single GPU in 1.5 days to 97% of the accuracy of agents trained on a prior state-of-the-art system.
  arXiv  Detail & Related papers  (2021-03-12T00:22:50Z)
• Sampling Training Data for Continual Learning Between Robots and the Cloud [26.116999231118793]
  We introduce HarvestNet, an intelligent sampling algorithm that resides on-board a robot and reduces system bottlenecks. It significantly improves the accuracy of machine-learning models on our novel dataset of road construction sites, field testing of self-driving cars, and streaming face recognition. It is between 1.05-2.58x more accurate than baseline algorithms and scalably runs on embedded deep learning hardware.
  arXiv  Detail & Related papers  (2020-12-12T05:52:33Z)
• Towards Scalable Distributed Training of Deep Learning on Public Cloud Clusters [30.4449309904155]
  We propose a new top-k sparsification communication library for distributed training. We show that our system achieves 25%-40% faster than existing state-of-the-art systems on CNNs and Transformer.
  arXiv  Detail & Related papers  (2020-10-20T17:16:29Z)
• One-step regression and classification with crosspoint resistive memory arrays [62.997667081978825]
  High speed, low energy computing machines are in demand to enable real-time artificial intelligence at the edge. One-step learning is supported by simulations of the prediction of the cost of a house in Boston and the training of a 2-layer neural network for MNIST digit recognition. Results are all obtained in one computational step, thanks to the physical, parallel, and analog computing within the crosspoint array.
  arXiv  Detail & Related papers  (2020-05-05T08:00:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.