t2i 开源项目推荐（49 个）

worldbench /DiffusionOPSD

🔥 On-Policy Self-Distillation in Diffusion Models

wildminder /ComfyUI-DyPE

ComfyUI DyPE+SEGA, enabling artifact-free 4K+ image generation: Z-Image, Qwen, Flux, Krea2

guyyariv /DyPE

[ICML 2026] Official implementation for "DyPE: Dynamic Position Extrapolation for Ultra High Resolution Diffusion".

aigem /aice_ps

Nano Banana editor(webui) for aistudio or gemini api key. 最强Ai修图APP，方便实用！ 如登录了google aistudio，可直接在DEMO APP中使用。 支持静态部署，兼容Gemini API

google /imageinwords

Data release for the ImageInWords (IIW) paper.

wildminder /awesome-qwen-image

Qwen-Image 2.1. Checkpoints, quants, prompt engines, LoRAs, and tooling

Osilly /Interleaving-Reasoning-Generation

[ICLR 2026] This is an early exploration to introduce Interleaving Reasoning to Text-to-image Generation field and achieve the SoTA benchmark performance. It also significantly improves the quality, fine-grained details and aesthetic aspects of generated images.

shawn0728 /Unify-Agent

🐧 Unify-Agent: An end-to-end unified multimodal agent for faithful, knowledge-grounded image generation.

blurgyy /CoMPaSS

[ICCV 2025] Enhancing spatial understanding in text-to-Image diffusion models

LeslieZhoa /Simple-Lora

diffusion lora chinese tutorial，虚拟idol训练中文教程

manh9011 /Perchance-T2I-Desktop

Desktop Client for Perchance Text to Image

moatifbutt /awesome-diffusion-iclr-2025

List of diffusion related active submissions on OpenReview for ICLR 2025.

OpenEnvision /AutoRubric-as-Reward

Auto-Rubric as Reward: From Implicit Preference to Explicit Generative Criteria

zhiyichin /P4D

[ICML 2024] Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Problematic Prompts (Official Pytorch Implementation)

Kaleidophon /token2index

A lightweight but powerful library to build token indices for NLP tasks, compatible with major Deep Learning frameworks like PyTorch and Tensorflow.

OpenRaiser /Envision

Envision: Benchmarking Unified Understanding & Generation for Causal World Process Insights

d60 /picwish

PicWish T2I, Photo Enhancer and Background Remover for Python

AMD-AGI /Instella-T2I

Instella-T2I: Pushing the Limits of 1D Discrete Latent Space Image Generation

MisterBrookT /IGenBench

[ACL 2026] A benchmark for evaluating the reliability of text-to-infographic generation with curated test cases and automated question-based verification.

VCG-team /elbo-t2ialign

(TIP 2026, CCF-A) ELBO-T2IAlign: A Generic ELBO-Based Method for Calibrating Pixel-level Text-Image Alignment in Diffusion Models