# Imagen 3

arXiv: https://arxiv.org/abs/2408.07009 · HTML: https://mlsanity.com/paper/2408.07009/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- First posted on arXiv 13 Aug 2024.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

## Built on by (most cited first, 27 of 27)

- [Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer](https://mlsanity.com/paper/2511.22699/) (2025)
- [TIIF-Bench: How Does Your T2I Model Follow Your Instructions?](https://mlsanity.com/paper/2506.02161/) (2025)
- [ACE-Step: A Step Towards Music Generation Foundation Model](https://mlsanity.com/paper/2506.00045/) (2025)
- [RynnVLA-001: Using Human Demonstrations to Improve Robot Manipulation](https://mlsanity.com/paper/2509.15212/) (2025)
- [UniVG: A Generalist Diffusion Model for Unified Image Generation and Editing](https://mlsanity.com/paper/2503.12652/) (2025)
- [OmnixR: Evaluating Omni-modality Language Models on Reasoning across Modalities](https://mlsanity.com/paper/2410.12219/) (2024)
- [Text-to-Image Diffusion Models Cannot Count, and Prompt Refinement Cannot Help](https://mlsanity.com/paper/2503.06884/) (2025)
- [HueManity: Probing Fine-Grained Visual Perception in MLLMs](https://mlsanity.com/paper/2506.03194/) (2025)
- [Exploiting Leaderboards for Large-Scale Distribution of Malicious Models](https://mlsanity.com/paper/2507.08983/) (2025)
- [Locality-Aware Continual Unlearning for Diffusion Models](https://mlsanity.com/paper/2512.02657/) (2025)
- [ChatDiT: A Training-Free Baseline for Task-Agnostic Free-Form Chatting with Diffusion Transformers](https://mlsanity.com/paper/2412.12571/) (2024)
- [HFMF: Hierarchical Fusion Meets Multi-Stream Models for Deepfake Detection](https://mlsanity.com/paper/2501.05631/) (2025)
- [Med-Art: Diffusion Transformer for 2D Medical Text-to-Image Generation](https://mlsanity.com/paper/2506.20449/) (2025)
- [Training-free Source Attribution of AI-generated Images via Resynthesis](https://mlsanity.com/paper/2510.24278/) (2025)
- [TexTailor: Inference-Time Textual Guidance Tailoring for Multimodal Diffusion Transformers](https://mlsanity.com/paper/2601.02211/) (2026)
- [Generalizing Score-based generative models for Heavy-tailed Distributions](https://mlsanity.com/paper/2603.00772/) (2026)
- [Exploring Gen-AI applications in building research and industry: A review](https://mlsanity.com/paper/2410.01098/) (2024)
- [WorldAgents: Can Foundation Image Models be Agents for 3D World Models?](https://mlsanity.com/paper/2603.19708/) (2026)
- [Non-Markov Multi-Round Conversational Image Generation with History-Conditioned MLLMs](https://mlsanity.com/paper/2601.20911/) (2026)
- [InpaintDPO: Mitigating Spatial Relationship Hallucinations in Foreground-conditioned Inpainting via Diverse Preference Optimization](https://mlsanity.com/paper/2512.15644/) (2025)
- [LivingWorld: Interactive 4D World Generation with Environmental Dynamics](https://mlsanity.com/paper/2604.01641/) (2026)
- [Embedding Arithmetic: A Lightweight, Tuning-Free Framework for Post-hoc Bias Mitigation in Text-to-Image Models](https://mlsanity.com/paper/2604.18167/) (2026)
- [BurstGP: Enhancing Raw Burst Image Super Resolution with Generative Priors](https://mlsanity.com/paper/2604.23508/) (2026)
- [DreamEdit3D: Personalization of Multi-View Diffusion Models for 3D Editing](https://mlsanity.com/paper/2605.16990/) (2026)
- [Training-Free Multi-Concept LoRA Composition with Prompt-Aware Weighting](https://mlsanity.com/paper/2606.03792/) (2026)
- [Diagnosing Aerial-View Object Detectors with Foundational Image Generative Models](https://mlsanity.com/paper/2607.02718/) (2026)
- [Layering Virtual Try-On](https://mlsanity.com/paper/2607.22924/) (2026)

## Abstract (arXiv)

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.