Effective Quantization of Muon Optimizer States

• URL: http://arxiv.org/abs/2509.23106v1
• Date: Sat, 27 Sep 2025 04:31:11 GMT
• Title: Effective Quantization of Muon Optimizer States
• Authors: Aman Gupta, Rafael Celente, Abhishek Shivanna, D. T. Braithwaite, Gregory Dexter, Shao Tang, Hiroto Udagawa, Daniel Silva, Rohan Ramanath, S. Sathiya Keerthi,
• Abstract summary: We introduce the 8-bit Muon using blockwise quantization, supporting both linear and dynamic schemes.<n>We demonstrate that 8-bit Muon maintains stability under both, while delivering $sim$74% reduction in memory footprint compared to full-precision Muon.<n>In extensive experiments, 8-bit Muon closely matches the performance of Muon while outperforming AdamW and 8-bit AdamW in pre-training a 1.6B model on 4B FineWeb tokens.
• Score: 6.256712531304834
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The Muon optimizer, based on matrix orthogonalization, has recently shown faster convergence and up to 2x computational efficiency over AdamW in LLM pretraining. Like AdamW, Muon is stateful, requiring storage of both model weights and accumulated gradients. While 8-bit AdamW variants mitigate this overhead using blockwise quantization, they are typically stable only under dynamic quantization - which improves stability on linear quantization for extreme values. In this paper, we introduce the 8-bit Muon optimizer using blockwise quantization, supporting both linear and dynamic schemes. We demonstrate that 8-bit Muon maintains stability under both, while delivering $\sim$74\% reduction in memory footprint compared to full-precision Muon. In extensive experiments, 8-bit Muon closely matches the performance of Muon while outperforming AdamW and 8-bit AdamW in pre-training a 1.6B model on 4B FineWeb tokens. It also shows competitive results when fine-tuning the Llama 3.2 3B model on post-training data. We also provide a theoretical perspective to help explain this robustness under quantization.

Related papers

• MuonBP: Faster Muon via Block-Periodic Orthogonalization [24.232069944820513]
  We show how to adjust the learning rate from the baseline to MuonBP and give guarantees for this algorithm.<n>When training an 8B model with eight-way tensor tensor and ZeRO statewiseing, MuonBP achieves 8% Muon with no degradation in performance.
  arXiv  Detail & Related papers  (2025-10-19T19:56:05Z)
• NorMuon: Making Muon more efficient and scalable [71.49702449498085]
  We propose NorMuon (Neuron-wise Normalized Muon) as a successor to Adam.<n>We show NorMuon consistently outperforms both Adam and Muon, achieving 21.74% better training efficiency than Adam and 11.31% improvement over Muon on 1.1 B pretraining setting.
  arXiv  Detail & Related papers  (2025-10-07T01:13:41Z)
• Binary Quantization For LLMs Through Dynamic Grouping [13.578307208515819]
  Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of Natural Language Processing (NLP) tasks.<n> Binary quantization, which compresses model weights from 16-bit Brain Float to 1-bit representations in -1, 1, offers significant reductions in storage and inference costs.<n>We propose a novel optimization objective tailored for binary quantization, along with three algorithms designed to realize it effectively.
  arXiv  Detail & Related papers  (2025-09-03T06:36:21Z)
• Pushing the Limits of Low-Bit Optimizers: A Focus on EMA Dynamics [64.62231094774211]
  Statefuls (e.g., Adam) maintain auxiliary information even 2x the model size in order to achieve optimal convergence.<n>SOLO enables Adam-styles to maintain quantized states with precision as low as 3 bits, or even 2 bits.<n>SOLO can thus be seamlessly applied to Adam-styles, leading to substantial memory savings with minimal accuracy loss.
  arXiv  Detail & Related papers  (2025-05-01T06:47:45Z)
• Muon is Scalable for LLM Training [50.68746986439438]
  We introduce Moonlight, a Mixture-of-Expert (MoE) model trained with 5.7T tokens using Muon.<n>Our model improves the current frontier, achieving better performance with much fewer training FLOPs compared to prior models.<n>We open-source our distributed Muon implementation that is memory optimal and communication efficient.
  arXiv  Detail & Related papers  (2025-02-24T09:12:29Z)
• ParetoQ: Scaling Laws in Extremely Low-bit LLM Quantization [58.84018707089315]
  We present a unified framework for rigorous comparisons across 1-bit, 1.58-bit, 2-bit, 3-bit, and 4-bit quantization settings.<n>We show that ternary, 2-bit, and 3-bit quantization maintains comparable performance in the size-accuracy trade-off.<n>Considering hardware constraints, 2-bit quantization offers promising potential for memory reduction and speedup.
  arXiv  Detail & Related papers  (2025-02-04T18:59:26Z)
• Analyzing Quantization in TVM [0.0]
  TVM has the ability to quantize weights and support low-bit computations. 8-bit quantization is usually expected to achieve around 50% of the full-precision inference time. In this project, we thoroughly investigate the reasons behind the underperformance and assess the compatibility and optimization opportunities of 8-bit quantization in TVM.
  arXiv  Detail & Related papers  (2023-08-19T07:39:46Z)
• 8-bit Optimizers via Block-wise Quantization [57.25800395197516]
  Statefuls maintain statistics over time, e.g., the exponentially smoothed sum (SGD with momentum) or squared sum (Adam) of past values. This state can be used to accelerate optimization compared to plain gradient descent but uses memory that might otherwise be allocated to model parameters. In this paper, we develop first gradients that use 8-bit statistics while maintaining the performance levels of using 32-bit gradient states.
  arXiv  Detail & Related papers  (2021-10-06T15:43:20Z)
• Leveraging Automated Mixed-Low-Precision Quantization for tiny edge microcontrollers [76.30674794049293]
  This paper presents an automated mixed-precision quantization flow based on the HAQ framework but tailored for the memory and computational characteristics of MCU devices. Specifically, a Reinforcement Learning agent searches for the best uniform quantization levels, among 2, 4, 8 bits, of individual weight and activation tensors. Given an MCU-class memory bound to 2MB for weight-only quantization, the compressed models produced by the mixed-precision engine result as accurate as the state-of-the-art solutions.
  arXiv  Detail & Related papers  (2020-08-12T06:09:58Z)

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.