Abra: Scaling Diffusion Image Training

• URL: http://arxiv.org/abs/2608.17286v1
• Date: Tue, 18 Aug 2026 02:36:25 GMT
• Title: Abra: Scaling Diffusion Image Training
• Abstract summary: We present a systematic scaling law study for text-to-image diffusion models.<n>We show that diffusion models scale just as predictably as language models but require far more data to train optimally.<n>We show that unlike language models, diffusion models are robust to overtraining and that practitioners should err on the side of more data rather than a larger model.
• Score: 19.77608057281383
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Compute-optimal scaling laws guide the training of frontier language models yet remain largely unexplored for visual generation. We present a systematic scaling law study for text-to-image diffusion models using Abra, a controlled family of flow-matching transformers trained across three orders of magnitude worth of compute ($10^{19}$ to $10^{22}$ FLOPs), reaching significantly larger compute budgets than previous works. We demonstrate that diffusion models scale just as predictably as language models but require far more data to train optimally: compute optimality occurs at approximately $200$ image tokens per parameter, ten times the Chinchilla compute-optimal prescription for LLMs. We show that unlike language models, diffusion models are robust to overtraining and that practitioners should err on the side of more data rather than a larger model. Finally, we show that this predictability extends beyond training loss to generative quality metrics, optimal CFG settings, representation quality, and even the shape of the training curves, which collapse onto a universal form.

Related papers

• How Far Can 5,500 Hours of Driving Take You? A Scaling Law Analysis of Video Diffusion Models [57.34034448977433]
  Video generation for autonomous driving is expensive to collect, bound by privacy requirements, and cannot be scraped at will.<n>We present a systematic scaling-law study of video diffusion models trained from scratch on driving data.<n>We train the largest video diffusion model trained from scratch on driving data: it sets a new open-source state of the art for driving video generation.
  arXiv  Detail & Related papers  (2026-08-28T14:56:01Z)
• Let's Scale Step by Step: Compute-Efficient Hyperparameter Transfer for Large-Scale Mixture-of-Experts [2.2554080765044717]
  Mixture-of-Experts (MoE) architectures significantly expand model capacity without a proportional increase in computational cost.<n>We propose a compute-efficient, two-step hyperparameter transfer framework that estimates optimal learning rates for training large MoE models.
  arXiv  Detail & Related papers  (2026-08-20T13:57:43Z)
• Scaling Law Analysis in Federated Learning: How to Select the Optimal Model Size? [12.791994483385409]
  Concerns are growing about the depletion of high-quality, well-curated training data.<n>The decentralization of training datasets in Federated Learning introduces challenges to scaling large models.<n>This paper provides insights on generalizing the previous model scaling experience to federated learning scenarios.
  arXiv  Detail & Related papers  (2025-11-15T12:41:25Z)
• Compute-Optimal Scaling for Value-Based Deep RL [99.680827753493]
  We investigate compute scaling for online, value-based deep RL.<n>Our analysis reveals a nuanced interplay between model size, batch size, and UTD.<n>We provide a mental model for understanding this phenomenon and build guidelines for choosing batch size and UTD.
  arXiv  Detail & Related papers  (2025-08-20T17:54:21Z)
• Predictable Scale: Part II, Farseer: A Refined Scaling Law in Large Language Models [62.3458061002951]
  We introduce Farseer, a novel and refined scaling law offering enhanced predictive accuracy across scales.<n>By systematically constructing a model loss surface $L(N,D)$, Farseer achieves a significantly better fit to empirical data than prior laws.<n>Our methodology yields accurate, robust, and highly generalizable predictions, demonstrating excellent extrapolation capabilities.
  arXiv  Detail & Related papers  (2025-06-12T17:59:23Z)
• Scaling Inference-Efficient Language Models [3.271571137474847]
  We show that model architecture affects inference latency, where models of the same size can have up to 3.5x difference in latency.<n>We modify the Chinchilla scaling laws to co-optimize the model parameter count, the number of training tokens, and the model architecture.<n>We release the Morph-1B model, which improves inference latency by 1.8x while maintaining accuracy on downstream tasks.
  arXiv  Detail & Related papers  (2025-01-30T03:16:44Z)
• Towards Precise Scaling Laws for Video Diffusion Transformers [43.6690970187664]
  We analyze scaling laws for video diffusion transformers and propose a new scaling law for any model size and compute budget.<n>Under these optimal settings, we achieve comparable performance and reduce inference costs by 40.1% compared to conventional scaling methods.
  arXiv  Detail & Related papers  (2024-11-25T18:59:04Z)
• More Compute Is What You Need [3.184416958830696]
  We propose a new scaling law that suggests model performance depends mostly on the amount of compute spent for transformer-based models. We predict that (a) for inference efficiency, training should prioritize smaller model sizes and larger training datasets, and (b) assuming the exhaustion of available web datasets, scaling the model size might be the only way to further improve model performance.
  arXiv  Detail & Related papers  (2024-04-30T12:05:48Z)
• FreeSeg-Diff: Training-Free Open-Vocabulary Segmentation with Diffusion Models [49.80911683739506]
  We focus on the task of image segmentation, which is traditionally solved by training models on closed-vocabulary datasets.<n>We leverage different and relatively small-sized, open-source foundation models for zero-shot open-vocabulary segmentation.<n>Our approach (dubbed FreeSeg-Diff), which does not rely on any training, outperforms many training-based approaches on both Pascal VOC and COCO datasets.
  arXiv  Detail & Related papers  (2024-03-29T10:38:25Z)
• Joint Adaptive Representations for Image-Language Learning [59.40890927221377]
  We propose a recipe for image-language learning, which produces effective models, outperforming bigger and more expensive ones, often trained on orders of magnitude larger datasets. Our key finding is the joint learning of a compact vision and language representation, which adaptively and iteratively fuses the multi-modal features. With only 40M training examples and with 39 GFLOPs our lightweight model outperforms many times larger state-of-the-art models of 2-20x more FLOPs and using bigger datasets some of which with close to 1B training examples.
  arXiv  Detail & Related papers  (2023-05-31T15:02:02Z)
• Ensembling Off-the-shelf Models for GAN Training [55.34705213104182]
  We find that pretrained computer vision models can significantly improve performance when used in an ensemble of discriminators. We propose an effective selection mechanism, by probing the linear separability between real and fake samples in pretrained model embeddings. Our method can improve GAN training in both limited data and large-scale settings.
  arXiv  Detail & Related papers  (2021-12-16T18:59:50Z)

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.