We propose Generalized Primal Averaging (GPA), an extension of Nesterov's method in its primal averaging formulation that addresses key limitations of recent averaging-based optimizers such as single-worker DiLoCo and Schedule-Free (SF) in the non-distributed setting. These two recent algorithmic approaches improve the performance of base optimizers, such as AdamW, through different iterate averaging strategies. Schedule-Free explicitly maintains a uniform average of past weights, while single-worker DiLoCo performs implicit averaging by periodically aggregating trajectories, called pseudo-gradients, to update the model parameters. However, single-worker DiLoCo's periodic averaging introduces a two-loop structure, increasing its memory requirements and number of hyperparameters. GPA overcomes these limitations by decoupling the interpolation constant in the primal averaging formulation of Nesterov. This decoupling enables GPA to smoothly average iterates at every step, generalizing and improving upon single-worker DiLoCo. Empirically, GPA consistently outperforms single-worker DiLoCo while removing the two-loop structure, simplifying hyperparameter tuning, and reducing its memory overhead to a single additional buffer. On the Llama-160M model, GPA provides a 24.22% speedup in terms of steps to reach the baseline (AdamW's) validation loss. Likewise, GPA achieves speedups of 12% and 27% on small and large batch setups, respectively, to attain AdamW's validation accuracy on the ImageNet ViT workload. Furthermore, we prove that for any base optimizer with regret bounded by $O(\sqrt{T})$, where $T$ is the number of iterations, GPA can match or exceed the convergence guarantee of the original optimizer, depending on the choice of interpolation constants.

Although LLM training is typically centralized with high-bandwidth interconnects and large compute budgets, emerging methods target communication-constrained training in distributed environments. The model trade-offs introduced by this shift remain underexplored, and our goal is to study them. We use the open-source nanochat project, a compact 8K-line full-stack ChatGPT-like implementation containing tokenization, pretraining, fine-tuning, and serving, as a controlled baseline. We implement the DiLoCo algorithm as a lightweight wrapper over nanochat's training loop, performing multiple local steps per worker before synchronization with an outer optimizer, effectively reducing communication by orders of magnitude. This inner-outer training is compared against a standard data-parallel (DDP) setup. Because nanochat is small and inspectable, it enables controlled pipeline adaptations and allows direct comparison with the conventional centralized baseline. DiLoCo achieves stable convergence and competitive loss in pretraining but yields worse MMLU, GSM8K, and HumanEval scores after mid-training and SFT. We discover that using DiLoCo-pretrained weights and running mid- and post-training with DDP fails to recover performance, revealing irreversible representation drift from asynchronous updates that impairs downstream alignment. We provide this implementation as an official fork of nanochat on GitHub.

Communication-efficient distributed training algorithms have received considerable interest recently due to their benefits for training Large Language Models (LLMs) in bandwidth-constrained settings, such as across datacenters and over the internet. Despite reducing communication frequency, these methods still typically require communicating a full copy of the model's gradients-resulting in a communication bottleneck even for cross-datacenter links. Furthermore, they can slightly degrade performance compared to a naive AdamW DDP baseline. While quantization is often applied to reduce the pseudo-gradient's size, in the context of LLM pre-training, existing approaches have been unable to additionally leverage sparsification and have obtained limited quantization. In this work, we introduce SparseLoCo, a communication-efficient training algorithm for LLMs that effectively leverages error feedback with Top-k sparsification and 2-bit quantization to reach extreme sparsity as low as 1-3% while outperforming full-precision DiLoCo. Our key observations are that outer momentum can be locally approximated by an error feedback accumulator combined with aggressive sparsity, and that sparse aggregation can actually improve model performance. We empirically demonstrate in a range of communication-constrained LLM training settings that SparseLoCo provides significant benefits in both performance and communication cost.

The rapid development of large language models (LLMs) has driven the demand for more efficient optimization techniques. Among these, the Lookahead family of optimizers employs a two-loop framework, maintaining fast and slow sets of model weights. Multiple inner optimizer steps on the fast weights produce a trajectory - the pseudo-gradient - that is used to update the slow weights. DiLoCo, a notable example originally designed for distributed training, applies Nesterov momentum to the averaged pseudo-gradient from multiple workers, claiming to even outperform AdamW in a non-distributed setup. In this paper, we empirically show that DiLoCo's surprising effectiveness stems primarily from applying Nesterov momentum to the pseudo-gradient, which improves training in a non-distributed setting. We call this Lookahead variant the Step-$K$ Nesterov Outer Optimizer (SNOO). We demonstrate that SNOO achieves compute factor gains of 1.5 - 2.5$\times$ in a non-distributed setting up to a scale of 1e23 training FLOPs, with improvements that increase with model size. Because of its minimal compute and memory overhead and compatibility with model sharding, SNOO is a practical enhancement for a variety of inner optimizers, including AdamW and Muon.

Following AI scaling trends, frontier models continue to grow in size and continue to be trained on larger datasets. Training these models requires huge investments in exascale computational resources, which has in turn driven developtment of distributed deep learning methods. Data parallelism is an essential approach to speed up training, but it requires frequent global communication between workers, which can bottleneck training at the largest scales. In this work, we propose a method called Pseudo-Asynchronous Local SGD (PALSGD) to improve the efficiency of data-parallel training. PALSGD is an extension of Local SGD (Stich, 2018) and DiLoCo (Douillard et al., 2023), designed to further reduce communication frequency by introducing a pseudo-synchronization mechanism. PALSGD allows the use of longer synchronization intervals compared to standard Local SGD. Despite the reduced communication frequency, the pseudo-synchronization approach ensures that model consistency is maintained, leading to performance results comparable to those achieved with more frequent synchronization. Furthermore, we provide a theoretical analysis of PALSGD, establishing its convergence and deriving its convergence rate. This analysis offers insights into the algorithm's behavior and performance guarantees. We evaluated PALSGD on image classification and language modeling tasks. Our results show that PALSGD achieves better performance in less time compared to existing methods like Distributed Data Parallel (DDP), and DiLoCo. Notably, PALSGD trains 18.4% faster than DDP on ImageNet-1K with ResNet-50, 24.4% faster than DDP on TinyStories with GPT-Neo-125M, and 21.1% faster than DDP on TinyStories with GPT-Neo-8M.

Scaling distributed training of Large Language Models (LLMs) requires not only algorithmic advances but also efficient utilization of heterogeneous hardware resources. While existing methods such as DiLoCo have demonstrated promising results, they often fail to fully exploit computational clusters under dynamic workloads. To address this limitation, we propose a three-stage method that combines Multi-Instance Training (MIT), Adaptive Batched DiLoCo, and switch mode mechanism. MIT allows individual nodes to run multiple lightweight training streams with different model instances in parallel and merge them to combine knowledge, increasing throughput and reducing idle time. Adaptive Batched DiLoCo dynamically adjusts local batch sizes to balance computation and communication, substantially lowering synchronization delays. Switch mode further stabilizes training by seamlessly introducing gradient accumulation once adaptive batch sizes grow beyond hardware-friendly limits. Together, these innovations improve both convergence speed and system efficiency. We also provide a theoretical estimate of the number of communications required for the full convergence of a model trained using our method.

Training large language models is generally done via optimization methods on clusters containing tens of thousands of accelerators, communicating over a high-bandwidth interconnect. Scaling up these clusters is expensive and can become impractical, imposing limits on the size of models that can be trained. Several recent studies have proposed training methods that are less communication intensive, avoiding the need for a highly connected compute cluster. These state-of-the-art low communication training methods still employ a synchronization step for model parameters, which, when performed over all model replicas, can become costly on a low-bandwidth network. In this work, we propose a novel optimization method, NoLoCo, that does not explicitly synchronize all model parameters during training and, as a result, does not require any collective communication. NoLoCo implicitly synchronizes model weights via a novel variant of the Nesterov momentum optimizer by partially averaging model weights with a randomly selected other one. We provide both a theoretical convergence analysis for our proposed optimizer as well as empirical results from language model training. We benchmark NoLoCo on a wide range of accelerator counts and model sizes, between 125M to 6.8B parameters. Our method requires significantly less communication overhead than fully sharded data parallel training or even widely used low communication training method, DiLoCo. The synchronization step itself is estimated to be one magnitude faster than the all-reduce used in DiLoCo for few hundred accelerators training over the internet. We also do not have any global blocking communication that reduces accelerator idling time. Compared to DiLoCo, we also observe up to $4\%$ faster convergence rate with wide range of model sizes and accelerator counts.

DiLoCo is a powerful framework for training large language models (LLMs) under networking constraints with advantages for increasing parallelism and accelerator utilization in data center settings. Despite significantly reducing communication frequency, however, DiLoCo's communication steps still involve all-reducing a complete copy of the model's parameters. While existing works have explored ways to reduce communication in DiLoCo, the role of error feedback accumulators and the effect of the inner-optimizer on compressibility remain under-explored. In this work, we investigate the effectiveness of standard compression methods including Top-k sparsification and quantization for reducing the communication overhead of DiLoCo when paired with two local optimizers (AdamW and Muon). Our experiments pre-training decoder-only transformer language models (LMs) reveal that leveraging Muon as the inner optimizer for DiLoCo along with an error-feedback accumulator allows to aggressively compress the communicated delta to 2-bits with next to no performance degradation. Crucially, MuLoCo (Muon inner optimizer DiLoCo) significantly outperforms DiLoCo while communicating 8X less and having identical memory complexity.

As we scale to more massive machine learning models, the frequent synchronization demands inherent in data-parallel approaches create significant slowdowns, posing a critical challenge to further scaling. Recent work develops an approach (DiLoCo) that relaxes synchronization demands without compromising model quality. However, these works do not carefully analyze how DiLoCo's behavior changes with model size. In this work, we study the scaling law behavior of DiLoCo when training LLMs under a fixed compute budget. We focus on how algorithmic factors, including number of model replicas, hyperparameters, and token budget affect training in ways that can be accurately predicted via scaling laws. We find that DiLoCo scales both predictably and robustly with model size. When well-tuned, DiLoCo scales better than data-parallel training with model size, and can outperform data-parallel training even at small model sizes. Our results showcase a more general set of benefits of DiLoCo than previously documented, including increased optimal batch sizes, improved downstream generalization with scale, and improved evaluation loss for a fixed token budget.

Distributed optimization methods such as DiLoCo have been shown to be effective in training very large models across multiple distributed workers, such as datacenters. These methods split updates into two parts: an inner optimization phase, where the workers independently execute multiple optimization steps on their own local data, and an outer optimization step, where the inner updates are synchronized. While such approaches require orders of magnitude less communication than standard data-parallel training, in settings where the workers are datacenters, even the limited communication requirements of these approaches can still cause significant slow downs due to the blocking necessary at each outer optimization step. In this paper, we investigate techniques to mitigate this issue by overlapping communication with computation in a manner that allows the outer optimization step to fully overlap with the inner optimization phase. We show that a particular variant, dubbed eager updates, provides competitive performance with standard DiLoCo in settings with low bandwidth between workers.