mlsanityData to 7 Oct 2026

Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer

arXiv 2511.22699 ↗First posted 27 Nov 2025cs.CVcited by 247

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

The landscape of high-performance image generation models is currently dominated by proprietary systems, such as Nano Banana Pro and Seedream 4.0. Leading open-source alternatives, including Qwen-Image, Hunyuan-Image-3.0 and FLUX.2, are characterized by massive parameter counts (20B to 80B), making them impractical for inference, and fine-tuning on consumer-grade hardware. To address this gap, we propose Z-Image, an efficient 6B-parameter foundation generative model built upon a Scalable Single-Stream Diffusion Transformer (S3-DiT) architecture that challenges the "scale-at-all-costs" paradigm. By systematically optimizing the entire model lifecycle -- from a curated data infrastructure to a streamlined training curriculum -- we complete the full training workflow in just 314K H800 GPU hours (approx. 630K). Our few-step distillation scheme with reward post-training further yields Z-Image-Turbo, offering both sub-second inference latency on an enterprise-grade H800 GPU and compatibility with consumer-grade hardware (<16GB VRAM). Additionally, our omni-pre-training paradigm also enables efficient training of Z-Image-Edit, an editing model with impressive instruction-following capabilities. Both qualitative and quantitative experiments demonstrate that our model achieves performance comparable to or surpassing that of leading competitors across various dimensions. Most notably, Z-Image exhibits exceptional capabilities in photorealistic image generation and bilingual text rendering, delivering results that rival top-tier commercial models, thereby demonstrating that state-of-the-art results are achievable with significantly reduced computational overhead. We publicly release our code, weights, and online demo to foster the development of accessible, budget-friendly, yet state-of-the-art generative models.

Key facts. Cited by 247 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 46 of its 100 references that are in the map. First posted on arXiv 27 Nov 2025.

Builds on

46 of 100 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Lumina-T2X: Transforming Text into Any Modality, Resolution, and Duration via Flow-based Large Diffusion Transformers9 May
  2. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis4 Jul

Other references · 44

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  2. 2025Qwen3 Technical Reportcited by 8,195
  3. 2022Flow Matching for Generative Modelingcited by 5,426
  4. 2025Qwen2.5-VL Technical Reportcited by 5,418
  5. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
  6. 2025SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Featurescited by 1,127
  7. 2025Qwen-Image Technical Reportcited by 973
  8. 2025FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Spacecited by 960
  9. 2025Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scalingcited by 787
  10. 2024Emu3: Next-Token Prediction is All You Needcited by 761
  11. 2023PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallelcited by 641
  12. 2025Flow-GRPO: Training Flow Matching Models via Online RLcited by 605
32 more
  1. 2024ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignmentcited by 455
  2. 2025Step1X-Edit: A Practical Framework for General Image Editingcited by 417
  3. 2025BLIP3-o: A Family of Fully Open Unified Multimodal Models-Architecture, Training and Datasetcited by 363
  4. 2024Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understandingcited by 294
  5. 2024Playground v2.5: Three Insights towards Enhancing Aesthetic Quality in Text-to-Image Generationcited by 249
  6. 2025Show-o2: Improved Native Unified Multimodal Modelscited by 242
  7. 2025OmniGen2: Towards Instruction-Aligned Multimodal Generationcited by 241
  8. 2025Seedream 4.0: Toward Next-generation Multimodal Image Generationcited by 233
  9. 2025ImgEdit: A Unified Image Editing Dataset and Benchmarkcited by 225
  10. 2025In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large Scale Diffusion Transformercited by 199
  11. 2024JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generationcited by 162
  12. 2024VisionReward: Fine-Grained Multi-Dimensional Human Preference Learning for Image and Video Generationcited by 148
  13. 2025Seedream 3.0 Technical Reportcited by 146
  14. 2025HunyuanImage 3.0 Technical Reportcited by 138
  15. 2022Chinese CLIP: Contrastive Vision-Language Pretraining in Chinesecited by 108
  16. 2025X-Omni: Reinforcement Learning Makes Discrete Autoregressive Image Generative Models Great Againcited by 105
  17. 2025Lumina-Image 2.0: A Unified and Efficient Image Generative Frameworkcited by 93
  18. 2025OneIG-Bench: Omni-dimensional Nuanced Evaluation for Image Generationcited by 81
  19. 2025Uniworld-V2: Reinforce Image Editing with Diffusion Negative-aware Finetuning and MLLM Implicit Feedbackcited by 78
  20. 2024CogView3: Finer and Faster Text-to-Image Generation via Relay Diffusioncited by 72
  21. 2025TIIF-Bench: How Does Your T2I Model Follow Your Instructions?cited by 64
  22. 2025UniWorld-V1: High-Resolution Semantic Encoders for Unified Visual Understanding and Generationcited by 64
  23. 2025Waver: Wave Your Way to Lifelike Video Generationcited by 52
  24. 2024Automatic Data Curation for Self-Supervised Learning: A Clustering-Based Approachcited by 48
  25. 2025Distribution Matching Distillation Meets Reinforcement Learningcited by 43
  26. 2025FLUX-Reason-6M & PRISM-Bench: A Million-Scale Text-to-Image Reasoning Dataset and Comprehensive Benchmarkcited by 37
  27. 2024Imagen 3cited by 26
  28. 2025Investigating Text Insulation and Attention Mechanisms for Complex Visual Text Generationcited by 22
  29. 2025LightGen: Efficient Image Generation through Knowledge Distillation and Direct Preference Optimizationcited by 20
  30. 20253DIS-FLUX: simple and efficient multi-instance generation with DiT renderingcited by 19
  31. 2025OmniCaptioner: One Captioner to Rule Them Allcited by 19
  32. 2025Factuality Matters: When Image Generation and Editing Meet Structured Visualscited by 16

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 243.

247 papers.

Sort:

By area:

  1. 2026DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Controlcited by 71
  2. 2026Helios: Real Real-Time Long Video Generation Modelcited by 56
  3. 2026ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?cited by 33
  4. 2026HiDream-O1-Image: A Natively Unified Image Generative Foundation Model with Pixel-level Unified Transformercited by 29
  5. 2026Qwen-Image-2.0 Technical Reportcited by 28
  6. 2026SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecturecited by 27
  7. 2026Gen-Searcher: Reinforcing Agentic Search for Image Generationcited by 25
  8. 2026InternVL-U: Democratizing Unified Multimodal Models for Understanding, Reasoning, Generation and Editingcited by 23
  9. 2026Mind-Brush: Integrating Agentic Cognitive Search and Reasoning into Image Generationcited by 18
  10. 2026Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generationcited by 17
  11. 2026D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Modelscited by 17
  12. 2026GEMS: Agent-Native Multimodal Generation with Memory and Skillscited by 16
38 more
  1. 2026Qwen-Image-Bench: From Generation to Creation in Text-to-Image Evaluationcited by 16
  2. 2026Think-Then-Generate: Reasoning-Aware Text-to-Image Diffusion with LLM Encoderscited by 15
  3. 2026GenEvolve: Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillationcited by 15
  4. 2026DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editingcited by 14
  5. 2026The Second Challenge on Real-World Face Restoration at NTIRE 2026: Methods and Resultscited by 13
  6. 2026Scaling Mixture-of-Experts Video Pretraining for Embodied Intelligencecited by 13
  7. 2026LLaDA2.0-Uni: Unifying Multimodal Understanding and Generation with Diffusion Large Language Modelcited by 12
  8. 2026DanceOPD: On-Policy Generative Field Distillationcited by 12
  9. 2026PromptRL: Prompt Matters in RL for Flow-Based Image Generationcited by 11
  10. 2026Thinking in Frames: How Visual Context and Test-Time Scaling Empower Video Reasoningcited by 10
  11. 2026DNA: Uncovering Universal Latent Forgery Knowledgecited by 10
  12. 2026The Fourth Challenge on Image Super-Resolution (×4) at NTIRE 2026: Benchmark Results and Method Overviewcited by 10
  13. 2026JoyAI-Image: Awaking Spatial Intelligence in Unified Multimodal Understanding and Generationcited by 10
  14. 2026GenClaw: Code-Driven Agentic Image Generationcited by 10
  15. 2026PISCO: Precise Video Instance Insertion with Sparse Controlcited by 9
  16. 2026Trust Your Critic: Robust Reward Modeling and Reinforcement Learning for Faithful Image Editing and Generationcited by 9
  17. 2026RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Modelscited by 9
  18. 2026CreatiParser: Generative Image Parsing of Raster Graphic Designs into Editable Layerscited by 8
  19. 2026Context Unrolling in Omni Modelscited by 8
  20. 2026Qwen-Image-Agent: Bridging the Context Gap in Real-World Image Generationcited by 8
  21. 2026TDM-R1: Reinforcing Few-Step Diffusion Models with Non-Differentiable Rewardcited by 7
  22. 2026BizGenEval: A Systematic Benchmark for Commercial Visual Content Generationcited by 7
  23. 2026L2P: Unlocking Latent Potential for Pixel Generationcited by 7
  24. 2026Aurora: Unified Video Editing with a Tool-Using Agentcited by 7
  25. 2026Spectrum Matching: a Unified Perspective for Superior Diffusability in Latent Diffusioncited by 6
  26. 2026PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusioncited by 6
  27. 2026Reinforcing Few-step Generators via Reward-Tilted Distribution Matchingcited by 6
  28. 2026Continuous Control of Editing Models via Adaptive-Origin Guidancecited by 5
  29. 2026Beyond VLM-Based Rewards: Diffusion-Native Latent Reward Modelingcited by 5
  30. 2026SceneAssistant: A Visual Feedback Agent for Open-Vocabulary 3D Scene Generationcited by 5
  31. 2026Visual Generation in the New Era: An Evolution from Atomic Mapping to Agentic World Modelingcited by 5
  32. 2026ERNIE-Image Technical Reportcited by 5
  33. 2026RS-Gen: A Multi-Stage Agentic Framework for Reasoning and Search-Augmented Image Generationcited by 5
  34. 2026COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflowscited by 5
  35. 2026OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformerscited by 5
  36. 2026Boogu-Image-0.1: Boosting Open Agentic Multimodal Generation via Understanding under a Minimal Budgetcited by 5
  37. 2025REVEAL: Reasoning-Enhanced Forensic Evidence Analysis for Explainable AI-Generated Image Detectioncited by 4
  38. 2026Semantic Routing: Exploring Multi-Layer LLM Feature Weighting for Diffusion Transformerscited by 4

50 most cited of 247 All 247 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026SnapGen++: Unleashing Diffusion Transformers for Efficient High-Fidelity Image Generation on Edge Devices
  2. 2026LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes
  3. 2026HRDiT: Training-Free High-Resolution Image Generation with Off-the-Shelf Diffusion Transformer Models
  4. 2024Delta-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers
  5. 2026Two Halves are More than One: Phase-wise Velocity Distillation for Fast and High-Quality Image Generation
  6. 2024PipeFusion: Patch-level Pipeline Parallelism for Diffusion Transformers Inference
  7. 2025DyDiT++: Diffusion Transformers with Timestep and Spatial Dynamics for Efficient Visual Generation
  8. 2025Toward Lightweight and Fast Decoders for Diffusion Models in Image and Video Generation