pith. sign in

arxiv: 2501.18427 · v4 · pith:3HE47GK5new · submitted 2025-01-30 · 💻 cs.CV

SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer

classification 💻 cs.CV
keywords scalingefficientmodelqualitysana-1computediffusiongeneration
0
0 comments X
read the original abstract

This paper presents SANA-1.5, a linear Diffusion Transformer for efficient scaling in text-to-image generation. Building upon SANA-1.0, we introduce three key innovations: (1) Efficient Training Scaling: A depth-growth paradigm that enables scaling from 1.6B to 4.8B parameters with significantly reduced computational resources, combined with a memory-efficient 8-bit optimizer. (2) Model Depth Pruning: A block importance analysis technique for efficient model compression to arbitrary sizes with minimal quality loss. (3) Inference-time Scaling: A repeated sampling strategy that trades computation for model capacity, enabling smaller models to match larger model quality at inference time. Through these strategies, SANA-1.5 achieves a text-image alignment score of 0.81 on GenEval, which can be further improved to 0.96 through inference scaling with VILA-Judge, establishing a new SoTA on GenEval benchmark. These innovations enable efficient model scaling across different compute budgets while maintaining high quality, making high-quality image generation more accessible. Our code and pre-trained models are released.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 26 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Flow-GRPO: Training Flow Matching Models via Online RL

    cs.CV 2025-05 unverdicted novelty 8.0

    Flow-GRPO is the first online RL method for flow matching models, raising GenEval accuracy from 63% to 95% and text-rendering accuracy from 59% to 92% with little reward hacking.

  2. OctoT2I: A Self-Evolving Agentic Text-to-Image Router

    cs.AI 2026-06 unverdicted novelty 7.0

    OctoT2I uses a no-supervision PSEL loop to discover model capability frontiers and route T2I tasks, reaching 0.96 GenEval score with 90.3% speedup over Flow-GRPO.

  3. The Silent Brush: Evaluating Artistic Style Leakage in AI Art Generation

    cs.LG 2026-05 unverdicted novelty 7.0

    Art Arena evaluates how artistic styles from training data leak into AI-generated images without explicit prompts, revealing asymmetric blending due to differences in representational strength and interaction dynamics...

  4. UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models

    cs.CV 2026-04 unverdicted novelty 7.0

    UDM-GRPO is the first RL integration for uniform discrete diffusion models, using final clean samples as actions and forward-process trajectory reconstruction to raise GenEval accuracy from 69% to 96% and OCR accuracy...

  5. In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large Scale Diffusion Transformer

    cs.CV 2025-04 unverdicted novelty 7.0

    ICEdit achieves state-of-the-art instructional image editing in Diffusion Transformers via in-context generation, requiring only 0.1% of prior training data and 1% trainable parameters.

  6. SciIR: A Large-scale Training Dataset and Benchmark for Scientific Image Reasoning Generation

    cs.CV 2026-06 unverdicted novelty 6.0

    Introduces SciIR-82k dataset and SciIR-Bench for scientific image reasoning generation organized by Peirce's semiotic triad, with fine-tuning raising model score from 35% to 43%.

  7. Don't Settle at the Mode! Mitigating Diversity Collapse in Pretrained Flow Models via Feature Self-Guidance

    cs.CV 2026-06 unverdicted novelty 6.0

    Feature self-guidance disperses internal features of flow models during batch generation and applies manifold regularization to increase output diversity while preserving condition alignment.

  8. IV-CoT: Implicit Visual Chain-of-Thought for Structure-Aware Text-to-Image Generation

    cs.CV 2026-06 unverdicted novelty 6.0

    IV-CoT introduces an implicit chain-of-thought framework that decomposes visual queries into a structural-to-semantic cascade with training-only sketch supervision to improve structure-aware text-to-image generation.

  9. FlowCodec: One-Step Flow Prior for Generative Image Compression

    eess.IV 2026-06 unverdicted novelty 6.0

    FlowCodec decouples latent compression from one-step flow-based transport to plug pretrained text-to-image models into ultra-low-bitrate codecs with under 0.54% trainable parameters.

  10. Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance

    cs.CV 2026-06 unverdicted novelty 6.0

    Moebius introduces a compressed diffusion inpainting model using Local-λ Mix Interaction blocks and latent-space multi-granularity distillation to reach 10B-level quality with 0.22B parameters.

  11. DuET: Dual Expert Trajectories for Diffusion Image Editing

    cs.CV 2026-06 unverdicted novelty 6.0

    DuET improves diffusion-based image editing by inserting a temporary text-to-image denoising phase to allow fuller execution of substantial edits while retaining structural guidance.

  12. SANA-Streaming: Real-time Streaming Video Editing with Hybrid Diffusion Transformer

    cs.CV 2026-05 unverdicted novelty 6.0

    SANA-Streaming delivers 1280x704 streaming video editing at 24 FPS end-to-end on an RTX 5090 using hybrid DiT blocks, cycle-reverse training, and mixed-precision quantization.

  13. What Makes Synthetic Data Effective in Image Segmentation

    cs.CV 2026-05 unverdicted novelty 6.0

    Dense scene composition and instance fidelity in synthetic diffusion images drive better segmentation performance; SENSE framework exploits this to improve models on Cityscapes, COCO, and ADE20K.

  14. When Policy Entropy Constraint Fails: Preserving Diversity in Flow-based RLHF via Perceptual Entropy

    cs.CV 2026-05 unverdicted novelty 6.0

    Policy entropy remains constant in flow-matching models during RLHF due to fixed noise schedules while perceptual diversity collapses from mode-seeking policy gradients, so perceptual entropy constraints are introduce...

  15. EPIC: Efficient Predicate-Guided Inference-Time Control for Compositional Text-to-Image Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    EPIC introduces predicate-guided inference-time search that lifts compositional T2I prompt accuracy from 34% to 71% on GenEval2 with 31-81% lower execution costs.

  16. UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models

    cs.CV 2026-04 unverdicted novelty 6.0

    UDM-GRPO integrates reinforcement learning with Uniform Discrete Diffusion Models by treating the final clean sample as the action and reconstructing trajectories via the forward process, achieving state-of-the-art te...

  17. Nucleus-Image: Sparse MoE for Image Generation

    cs.CV 2026-04 unverdicted novelty 6.0

    A 17B-parameter sparse MoE diffusion transformer activates 2B parameters per pass and reaches competitive quality on image generation benchmarks without post-training.

  18. TC-AE: Unlocking Token Capacity for Deep Compression Autoencoders

    cs.CV 2026-04 unverdicted novelty 6.0

    TC-AE improves reconstruction and generative performance in deep compression by decomposing token-to-latent compression into two stages and using joint self-supervised training.

  19. RAPO++: Cross-Stage Prompt Optimization for Text-to-Video Generation via Data Alignment and Test-Time Scaling

    cs.CV 2025-10 unverdicted novelty 6.0

    RAPO++ is a three-stage prompt optimization framework combining retrieval-augmented refinement, closed-loop test-time scaling, and LLM fine-tuning to enhance text-to-video generation quality.

  20. Large Scale Diffusion Distillation via Score-Regularized Continuous-Time Consistency

    cs.CV 2025-10 conditional novelty 6.0

    The work introduces rCM, a score-regularized continuous-time consistency model that matches DMD2 quality on large models up to 14B parameters while improving diversity and enabling 1-4 step sampling.

  21. YOLOv12: Attention-Centric Real-Time Object Detectors

    cs.CV 2025-02 unverdicted novelty 6.0

    YOLOv12 is a new attention-based real-time object detector that reports higher accuracy than YOLOv10, YOLOv11, and RT-DETR variants at comparable or better speed and efficiency.

  22. TMP: Tree-structured Mixed-policy Pruning for Large-scale Image Generation and Editing

    cs.CV 2026-06 unverdicted novelty 5.0

    TMP is a pruning framework that reduces HunyuanImage-3.0 from 80B to 20B parameters (75% reduction) and Z-Image turbo from 6B to 4B with limited quality degradation.

  23. Test-Time Scaling in Multimodal Foundation Models: A Comprehensive Survey of Generation and Reasoning

    cs.CV 2026-06 unverdicted novelty 5.0

    A survey of test-time scaling for multimodal foundation models that introduces a three-way taxonomy of sampling, feedback, and search approaches along with applications and benchmarks.

  24. WinTok: A Win-Win Hybrid Tokenizer via Decomposing Visual Understanding and Generation with Transferable Tokens

    cs.CV 2026-05 unverdicted novelty 5.0

    WinTok is a hybrid visual tokenizer that supplements pixel tokens with learnable semantic tokens distilled asymmetrically from foundation models to improve reconstruction, understanding, and generation.

  25. Qwen-Image Technical Report

    cs.CV 2025-08 unverdicted novelty 5.0

    Qwen-Image is a foundation model that reaches state-of-the-art results in image generation and editing by combining a large-scale text-focused data pipeline with curriculum learning and dual semantic-reconstructive en...

  26. Breaking the Lock-in: Diversifying Text-to-Image Generation via Representation Modulation

    cs.CV 2026-06 unverdicted novelty 4.0

    Early DC component convergence in text-to-image Transformer features causes output homogeneity; selective early attenuation via DAVE improves diversity without retraining or extra cost.