BarretBa

BeyondTimeShifts. Beyond Time Shifts (ECCV 2026): a reference-free audio-visual synchronization evaluator built on Qwen2.5-Omni-3B, with SynthSync data, R-GRPO training, and the SyncBench benchmark.

JavisDiT. [ICLR 2026] Official implementation of JavisDiT and JavisDiT++ series.

r1_reward. ✨✨ [ICLR 2026] R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement Learning

OmniReward. Python

LLaVA-Hound-DPO. Python

Omni-RRM. Python

perception_models. State-of-the-art Image & Video CLIP, Multimodal Large Language Models, and More!

TalkVid. TalkVid: A Large-Scale Diversified Dataset for Audio-Driven Talking Head Synthesis [CVPR 2026 Findings]

DreamOmni2. This project is the official implementation of 'DreamOmni2: Multimodal Instruction-based Editing and Generation (CVPR2026 Highlight)''

insightface. State-of-the-art 2D and 3D Face Analysis Project

JoyAI-Echo. JoyAI-Echo: Pushing the Frontier of Long Audio-Visual Generation

DreamID-Omni. [ICML 2026] DreamID-Omni: Unified Framework for Controllable Human-Centric Audio-Video Generation

Wan2.2. Wan: Open and Advanced Large-Scale Video Generative Models

Audio-Omni. [SIGGRAPH 2026] Repository of Audio-Omni

ID-LoRA. [ECCV 2026] Generate high resolution videos with a custom voice and appearance, based on LTX-2/LTX-2.3 + Identity In-Context LoRA

OmniCustom. Official Implementation of 'OmniCustom: Sync Audio-Video Customization Via Joint Audio-Video Generation Model'

Phantom. Phantom: Subject-Consistent Video Generation via Cross-Modal Alignment

Identity-as-Presence. Python

Qwen3-Omni. Qwen3-omni is a natively end-to-end, omni-modal LLM developed by the Qwen team at Alibaba Cloud, capable of understanding text, audio, images, and video, as well as generating speech in real time.

ConsisID. [CVPR 2025 Highlight🔥] Identity-Preserving Text-to-Video Generation by Frequency Decomposition

ROLL. An Efficient and User-Friendly Scaling Library for Reinforcement Learning with Large Language Models

al-folio. A beautiful, simple, clean, and responsive Jekyll theme for academics

mellea. Mellea is a library for writing generative programs.

Awesome-Video-Generation-Post-Training. [TMLR] Video Generation Models: A Survey of Post-Training and Alignment | 🔥 A continuously updated collection of papers, datasets, and benchmarks on post-training and alignment for video generation.

python3-cookbook. 《Python Cookbook》 3rd Edition Translation

DiffusionNFT. [ICLR 2026 Oral] DiffusionNFT: Online Diffusion Reinforcement with Forward Process

TIIF-Bench. Official repository for the paper "TIIF-Bench: How Does Your T2I Model Follow Your Instructions?".

OneIG-Benchmark. [NeurIPS 2025 DB] OneIG-Bench is a meticulously designed comprehensive benchmark framework for fine-grained evaluation of T2I models across multiple dimensions, including subject-element alignment, text rendering precision, reasoning-generated content, stylization, and diversity.

VACE. [ICCV 2025] Official implementations for paper: VACE: All-in-One Video Creation and Editing

SpatialT2I. [CVPR 2026🔥] Enhancing Spatial Understanding in Image Generation via Reward Modeling

Pref-GRPO. Official implementation of Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning

Flow-Factory. A unified framework for easy reinforcement learning in Flow-Matching models

EditReward. EditReward: A Human-Aligned Reward Model for Instruction-Guided Image Editing [ICLR 2026]

UnifiedReward. Official implementation of UnifiedReward & [NeurIPS 2025] UnifiedReward-Think & UnifiedReward-Flex

InternVL. [CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的开源多模态对话模型

VLM-R1. Solve Visual Understanding with Reinforced VLMs

UniPercept. [ICML2026 Spotlight] UniPercept: Towards Unified Perceptual-Level Image Understanding across Aesthetics, Quality, Structure, and Texture

LlamaFactory. Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

lightly-studio. Curate, Annotate, and Manage Your Data in LightlyStudio.

Awesome-Unified-Multimodal-Models. Awesome Unified Multimodal Models

CLIPood. About Code Release for "CLIPood: Generalizing CLIP to Out-of-Distributions" (ICML 2023), https://arxiv.org/abs/2302.00864

SRPO. Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference

DanceGRPO. An official implementation of DanceGRPO: Unleashing GRPO on Visual Generation

Qwen3-VL. Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

Qwen-Image. Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.

GLM-V. GLM-4.6V/4.5V/4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning

learn-user-pref. Official implementation of "Learning User Preferences for Image Generation Models"

MoMu. The PyTorch implementation of MoMu, described in "Natural Language-informed Modeling of Molecule Graphs".

Enformer_Borzoi_Training_Pytorch. We provide pytorch training scripts for enformer and borzoi

aesthetic-predictor-v2-5. SigLIP-based Aesthetic Score Predictor

SPACE. Python

MindtheGap. ICCV25 highlight

acad-homepage.github.io. AcadHomepage: A Modern and Responsive Academic Personal Homepage

DAS. Official implementation for Diffusion Alignment as Sampling (DAS), ICLR'25, Spotlight

Bagel. Open-source unified multimodal model

flow_grpo. [NeurIPS 2025] An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL

ddpo. Code for the paper "Training Diffusion Models with Reinforcement Learning"

DeepSeek-Math. DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Dual-Diffusion. Code for D-DiT

ParetoFlow. [ICLR 2025] Official Implementation of ParetoFlow: Guided Flows in Multi-Objective Optimization🧬🧬🧬

Awesome-Multi-Objective-Deep-Learning. A comprehensive list of gradient-based multi-objective optimization algorithms in deep learning.

Janus. Janus-Series: Unified Multimodal Understanding and Generation Models

latent-consistency-model. Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference

rg-lcd. Reward Guided Latent Consistency Distillation

MetaMask. Python

CPKP. Python

Uniform-Attention-Maps. [WACV 2025] Uniform Attention Maps: Enhancing Image Fidelity in Reconstruction and Editing

Reward-Instruct. [NeurIPS 2025] Reward-Instruct: A Reward-Centric Approach to Fast Photo-Realistic Image Generation

CrossFlow. [CVPR2025] PyTorch-based reimplementation of CrossFlow, as proposed in 'Flowing from Words to Pixels: A Noise-Free Framework for Cross-Modality Evolution'

transfusion-pytorch. Pytorch implementation of Transfusion, "Predict the Next Token and Diffuse Images with One Multi-Modal Model", from MetaAI

mar. PyTorch implementation of MAR+DiffLoss https://arxiv.org/abs/2406.11838

TACO. TACO: TFBS-Aware Cis-Regulatory Element Optimization

ProtDETR. Interpretable Enzyme Function Prediction via Residue-Level Detection

newPCMDM. Python

PCMDM. Official Implementation of "Synthesizing Long-Term Human Motions with Diffusion Models via Coherent Sampling"

GraphTextRetrieval. Part of official implementation of "Natural language-informed learning of molecule graphs"

InstantStyle. InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation 🔥

ZegCLIP. Official implement of CVPR2023 ZegCLIP: Towards Adapting CLIP for Zero-shot Semantic Segmentation

Fk-Diffusion-Steering. A general framework for inference-time scaling and steering of diffusion models with arbitrary rewards.

SimpleTuner. A general fine-tuning kit geared toward image/video/audio diffusion models.

geneval. GenEval: An object-focused framework for evaluating text-to-image alignment

AlignProp. AlignProp uses direct reward backpropogation for the alignment of large-scale text-to-image diffusion models. Our method is 25x more sample and compute efficient than reinforcement learning methods (PPO) for finetuning Stable Diffusion

PickScore. Python

BLIP. PyTorch code for BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation

ImageReward. [NeurIPS 2023] ImageReward: Learning and Evaluating Human Preferences for Text-to-image Generation

open-prompts. Python

textcraftor. Python

TexForce. Official PyTorch codes for "Enhancing Diffusion Models with Text-Encoder Reinforcement Learning", ECCV2024

ELLA. ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment

text-to-text-transfer-transformer. Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"

T2I-CompBench. [Neurips 2023 & TPAMI] T2I-CompBench (++) for Compositional Text-to-image Generation Evaluation

sd3.5. Python

ReNO. [NeurIPS 2024] ReNO: Enhancing One-step Text-to-Image Models through Reward-based Noise Optimization

Structured-Diffusion-Guidance. Training-Free Structured Diffusion Guidance for Compositional Text-to-Image Synthesis

CogVLM2. GPT4V-level open-source multi-modal model based on Llama3-8B

Open-O1. Python

open-o1. open-o1: Using GPT-4o with CoT to Create o1-like Reasoning Chains

mm-cot. Official implementation for "Multimodal Chain-of-Thought Reasoning in Language Models" (stay tuned and more will be updated)

aesthetic-predictor. A linear estimator on top of clip to predict the aesthetic quality of pictures

NExT-GPT. Code and models for ICML 2024 paper, NExT-GPT: Any-to-Any Multimodal Large Language Model

RPG-DiffusionMaster. [ICML 2024] Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs (RPG)

PMG. The repository of paper Personalized Multimodal Response Generation with Large Language Models

Vitron. NeurIPS 2024 Paper: A Unified Pixel-level Vision LLM for Understanding, Generating, Segmenting, Editing

long_stable_diffusion. Long-form text-to-images generation, using a pipeline of deep generative models (GPT-3 and Stable Diffusion)

SEED. Official implementation of SEED-LLaMA (ICLR 2024).

gill. 🐟 Code and models for the NeurIPS 2023 paper "Generating Images with Multimodal Language Models".

HPSv2. Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis

ComfyUI. The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.

flux. Application Architecture for Building User Interfaces

DEADiff. [CVPR 2024] Official implementation of "DEADiff: An Efficient Stylization Diffusion Model with Disentangled Representations"

google-research. Google Research

daam. Diffusion attentive attribution maps for interpreting Stable Diffusion.

stable-diffusion-webui-daam. DAAM for Stable Diffusion Web UI

LLaVA. [NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

Attend-and-Excite. Official Implementation for "Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models" (SIGGRAPH 2023)

training-free-structured-diffusion-guidance. 🤗 Unofficial huggingface/diffusers-based implementation of the paper "Training-Free Structured Diffusion Guidance for Compositional Text-to-Image Synthesis".

SeedSelect. Code for our papers : "Generating images of rare concepts using pre-trained diffusion models" (AAAI 24) and "Norm-guided latent space exploration for text-to-image generation" (Neurips 23)

StyleDrop-PyTorch. Unoffical implement for [StyleDrop](https://arxiv.org/abs/2306.00983)

sd-webui-fabric. Python

fabric. Python

initno. [CVPR 2024] InitNO: Boosting Text-to-Image Diffusion Models via Initial Noise Optimization

direct-preference-optimization. Reference implementation for DPO (Direct Preference Optimization)

Open-Sora. Open-Sora: Democratizing Efficient Video Production for All

CogVideo. text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

MCTS-DPO. This is the repository that contains the source code for the Self-Evaluation Guided MCTS for online DPO.

ziplora-pytorch. Implementation of "ZipLoRA: Any Subject in Any Style by Effectively Merging LoRAs"

SPO. [CVPR 2025] Aesthetic Post-Training Diffusion Models from Generic Preferences with Step-by-step Preference Optimization

mapo. Official codebase for Margin-aware Preference Optimization for Aligning Diffusion Models without Reference (MaPO).

VADER. Video Diffusion Alignment via Reward Gradients. We improve a variety of video diffusion models such as VideoCrafter, OpenSora, ModelScope and StableVideoDiffusion by finetuning them using various reward models such as HPS, PickScore, VideoMAE, VJEPA, YOLO, Aesthetics etc.

d3po. [CVPR 2024] Code for the paper "Using Human Feedback to Fine-tune Diffusion Models without Any Reward Model"

DiffusionDPO. Code for "Diffusion Model Alignment Using Direct Preference Optimization"

Free-Guidance-Diffusion. A novel method that provides greater control over generated images by guiding the internal representations of the pre-trained Stable Diffusion.

PAE. [CVPR 2024] Dynamic Prompt Optimizing for Text-to-Image Generation

LLaMA2-Accessory. An Open-source Toolkit for LLM Development

IP-Adapter. The image prompt adapter is designed to enable a pretrained text-to-image diffusion model to generate images with image prompt.

Dreambooth-Stable-Diffusion. Implementation of Dreambooth (https://arxiv.org/abs/2208.12242) by way of Textual Inversion (https://arxiv.org/abs/2208.01618) for Stable Diffusion (https://arxiv.org/abs/2112.10752). Tweaks focused on training faces, objects, and styles.

diffusers. 🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.

Dreambooth-Stable-Diffusion. Implementation of Dreambooth (https://arxiv.org/abs/2208.12242) with Stable Diffusion

CLIP-Adapter. Python

Transfer-Learning-Library. Transfer Learning Library for Domain Adaptation, Task Adaptation, and Domain Generalization

BatchFormer. CVPR2022, BatchFormer: Learning to Explore Sample Relationships for Robust Representation Learning, https://arxiv.org/abs/2203.01522

APE. [ICCV 2023] Code for "Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinement"

TPT. Test-time Prompt Tuning (TPT) for zero-shot generalization in vision-language models (NeurIPS 2022))

CLIP. CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

ICLR24. Official code for ICLR 2024 paper, "A Hard-to-Beat Baseline for Training-free CLIP-based Adaptation"

SuS-X. Code for the paper: "SuS-X: Training-Free Name-Only Transfer of Vision-Language Models" [ICCV'23]

Tip-Adapter. Python

DPT. Official PyTorch implementation for "Diffusion Models and Semi-Supervised Learners Benefit Mutually with Few Labels"

VAEs. Variational autoencoders: VAE, gaussian mixture VAE (GMVAE), and a basic ladder VAE (LVAE)

AugSelf. Python

self-supervised. Whitening for Self-Supervised Representation Learning | Official repository

DomainBed. DomainBed is a suite to test domain generalization algorithms

lightly. A python library for self-supervised learning on images.

LINKX_paddle. Unofficial PaddlePaddle Implementation of LINKX (Large Scale Learning on Non-Homophilous Graphs)