Junyan ye

superpowers. An agentic skills framework & software development methodology that works.

pelican-bicycle. LLM benchmark: Generate an SVG of a pelican riding a bicycle

code_aesthetics. Official repository for paper: Code Aesthetics with Agentic Reward Feedback

DesignBench. DesignBench: A Comprehensive Benchmark for MLLM-based Front-end Code Generation

Fast-dLLM. Official implementation of "Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding"

open-design. 🎨 The open-source Claude Design alternative. 🖥️ Local-first desktop app. 🖼️ Your coding agent becomes the design engine: prototypes, landing pages, dashboards, slides, images & video — real files, HTML/PDF/PPTX/MP4 export. 🤖 Claude Code / Codex / Cursor / Gemini / OpenCode / Qwen & 20+ CLIs via BYOK.

ArtifactsBenchmark. Python

PosterCopilot.

pixel2motion. AI logo animation skill: turn raster logos into smooth SVG animation, animated HTML demos, GIF/video previews, and motion QA evidence.

manim. Animation engine for explanatory math videos

Code2Video. [ICML 2026] Video generation via code

LASEV. HTML

ViMax. "ViMax: Agentic Video Generation (Director, Screenwriter, Producer, and Video Generator All-in-One)"

hyperframes. Write HTML. Render video. Built for agents.

seedance2.0-prompt-skill.

baoyu-skills. TypeScript

AgentGym-RL. Code and implementations for the paper "AgentGym-RL: Training LLM Agents for Long-Horizon Decision Making through Multi-Turn Reinforcement Learning" by Zhiheng Xi et al.

SandboxFusion. Python

ToRL. Python

meta-skill. meta skill — agents help you post demos and learn from each other at Skill Evolve

OmniBehavior. Code for "Towards Real-world Human Behavior Simulation: Benchmarking Large Language Models on Long-horizon, Cross-scenario, Heterogeneous Behavior Traces", where dataset is publicly available at https://huggingface.co/datasets/jiawei-ucas/OmniBehavior

claude-plugins-official. Official, Anthropic-managed directory of high quality Claude Code Plugins.

OpenSkill. Open-World Self-Evolution for LLM Agents — agents that build both their skills and their own verification signals from scratch, with no target-task supervision. (Code coming soon.)

photodesign-skill. 一个专业的摄影场景设计技能skill，用于图像模型生图或编辑。

Memento-Skills. Memento-Skills: Let Agents Design Agents

awesome-claude-skills. A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows

ralph. Ralph is an autonomous AI agent loop that runs repeatedly until all PRD items are complete.

AgentRL. Scaling Agentic Reinforcement Learning with a Multi-Turn, Multi-Task Framework

Qwen-AgentWorld. Qwen-AgentWorld: Language World Models for General Agents

ReAct. [ICLR 2023] ReAct: Synergizing Reasoning and Acting in Language Models

OpenThoughts-Agent. Data recipes and robust infrastructure for training AI agents

SWE-agent. SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]

SWE-bench. SWE-bench: Can Language Models Resolve Real-world Github Issues?

guizang-social-card-skill. 🪧 Claude Code / Codex skill — generate Xiaohongshu carousels & WeChat 21:9+1:1 cover pairs. Editorial × Swiss visual systems, 28 layouts, 10 themes, single-file HTML → PNG. 小红书图文 + 公众号封面对

agent-mangaka-forge. AI manga page generation skill pack with persistent character image memory for consistent characters across pages.

bpy-renderer. Python package for rendering 3D scenes and animations using blender.

image-extender. TypeScript

SkillOpt. SkillOpt is a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts.

cosmos. NVIDIA Cosmos is an open platform of world models, datasets, and tools that enables developers to build Physical AI for robots, autonomous vehicles, smart infrastructure, and more.

ideogram4. Ideogram 4: Open image model at the forefront of design

GPT-Image2-Skill. GPT Image 2 prompt gallery, image prompt library, agentic skill, and CLI for OpenAI image generation/editing

agent-banana.

colleague-skill. 将冰冷的离别化为温暖的 Skill，欢迎加入数字生命1.0！Transforming cold farewells into warm skills? It's giving rebirth era. Welcome to Digital Life 1.0. 🫶

count-anything. Code and implementation guidelines for the paper Counting Anything. Project Page: https://mengqi-lei.github.io/count-anything-projectpage/

magic-agents. Agent graph builder

GenClaw. GenClaw: Code-Driven Agentic Image Generation

AutoSkill. AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

a-evolve. The official repository of "Position: Agentic Evolution is the Path to Evolving LLMs".

Gen-Searcher. Gen-Searcher: Reinforcing Agentic Search for Image Generation

Awesome-Code-as-Agent-Harness-Papers. A curated list of papers and resources based on the survey "Code as Agent Harness"

GenEvolve. Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillation

pi. AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

SCOPE. Python

articraft. An Agentic System for Scalable Articulated 3D Asset Generation

Lance. A 3B-active-parameter native unified multimodal model for image and video understanding, generation, and editing.

Thinking-with-Visual-Primitives-pytorch. Unofficial PyTorch reproduction of DeepSeek's Thinking with Visual Primitives.

D-OPSD. Official Repo of "D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models"

Thinking-with-Visual-Primitives. Clone of DeepSeek Thinking-with-Visual-Primitives

SenseNova-U1. SenseNova-U series: Native Unified Paradigm with NEO-unify from the First Principles

tldraw. Build infinite canvas apps in React with the tldraw SDK. World's best, top-most agent recommended #1 five star SDK.

awesome-gpt-image-2. 🚀 World's largest GPT Image 2 prompt library, updated daily — 2000+ curated prompts with preview images, 16 languages. OpenAI's next-gen image model with pixel-perfect text rendering, cross-image consistency, and commercial-grade illustration. Free & open source.

excalidraw-mcp. Fast and streamable Excalidraw MCP App

verl-agent. verl-agent is an extension of veRL, designed for training LLM/VLM agents via RL. verl-agent is also the official code for paper "Group-in-Group Policy Optimization for LLM Agent Training"

ClawGUI. Build, Evaluate, and Deploy GUI Agents — online RL training, standardized benchmarks, and real-device deployment in one framework.

LaviGen. [CVPR 2026] Official Implementation of "Repurposing 3D Generative Model for Autoregressive Layout Generation"

EUPE. Efficient Universal Perception Encoder: a single on-device vision encoder with versatile representations that match or exceed specialized experts across multiple task domains.

Awesome-Remote-Sensing-Agents. 🚀Official Repository of Intelligent Remote Sensing Agents: A Survey

oracle. Ask the oracle when you're stuck. Invoke GPT-5 Pro with a custom context and files.

HiVG. Hierarchical SVG Tokenization: Learning Compact Visual Programs for Scalable Vector Graphics Modeling

PSDesigner. [CVPR 2026] PSDesigner: Automated Graphic Design with a Human-Like Creative Workflow

JarvisEvo. [CVPR' 2026] JarvisEvo: Towards a Self-Evolving Photo Editing Agent with Synergistic Editor-Evaluator Optimization

mulan-code. Python

SciSketch. Python

PosterMaker. PosterMaker [CVPR 2025] https://poster-maker.github.io/

Rise-Video. RISE-Video: Can Video Generators Decode Implicit World Rules?

JoyAI-Image. JoyAI-Image is the unified multimodal foundation model for image understanding, text-to-image generation, and instruction-guided image editing.

vtracer. Raster to Vector Graphics Converter

InternSVG. [ICLR 2026] Official repository of "InternSVG: Towards Unified SVG Tasks with Multimodal Large Language Models".

FIPO. This code implements the algorithm of FIPO, a value-free RL recipe for eliciting deeper reasoning from a clean base model.

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

Awesome-Generative-Models-for-OCR. [arXiv 25] OCRGenBench: A Comprehensive Benchmark for Evaluating OCR Generative Capabilities

VCode. VCode: SVG as Symbolic Visual Representation

OverLayBench. (NeurIPS 2025 D&B Track) OverLayBench: A Benchmark for Layout-to-Image Generation with Dense Overlaps

WeEdit. A Dataset, Benchmark and Glyph-Guided Framework for Text-centric Image Editing

Code2World. Code2World: A GUI World Model via Renderable Code Generation

gWorld. Generative Visual Code Mobile World Model

CTRL-S. [ECCV 2026] Official repository of "Reliable Reasoning in SVG-LLMs via Multi-Task Multi-Reward Reinforcement Learning".

OmniLottie. [CVPR 2026🔥] 🧑‍🎨 OmniLottie, an open-sourced multi-modal instructed vector animation generator that produces Lottie JSONs.

OmniSVG. [NeurIPS 2025] OmniSVG is the first family of end-to-end multimodal SVG generators that leverage pre-trained Vision-Language Models (VLMs), capable of generating complex and detailed SVGs, from simple icons to intricate anime characters.

UniRef. [ICCV2023] Segment Every Reference Object in Spatial and Temporal Spaces

Understand-Anything. Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.

star-vector. StarVector is a foundation model for SVG generation that transforms vectorization into a code generation task. Using a vision-language modeling architecture, StarVector processes both visual and textual inputs to produce high-quality SVG code with remarkable precision.

AutoFigure-Edit. Python

Structured-Visuals. [ICLR2026] Factuality Matters: When Image Generation and Editing Meet Structured Visuals

CoCo. CoCo: Code as CoT for Text-to-Image Preview and Rare Concept Generation

Fibo-Edit. FIBO-Edit brings the power of structured prompt generation to image editing

sam2. The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.

TDM-R1. [ICML 2026][Ultra Powerful Few-Step Diffusion RL] TDM-R1: Reinforcing Few-Step Diffusion Models with Non-Differentiable Reward

JiT. PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

Auto-claude-code-research-in-sleep. ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

ComfyUI-EditUtils. a set of utils for comfyui edit model

Comfyui-QwenEditUtils. Python

OpenClaw-RL. OpenClaw-RL: Train any agent simply by talking

mmagic. OpenMMLab Multimodal Advanced, Generative, and Intelligent Creation Toolbox. Unlock the magic 🪄: Generative-AI (AIGC), easy-to-use APIs, awsome model zoo, diffusion models, for text-to-image generation, image/video restoration/enhancement, etc.

KV-Edit. [ICCV 2025] Official implementation for KV-Edit: Training-Free Image Editing for Precise Background Preservation

CoCoEdit. [ICML 2026] Official PyTorch implementation of paper “CoCoEdit: Content-Consistent Image Editing via Region Regularized Reinforcement Learning”

GVCoT.

FireRed-Image-Edit. FireRed-Image-Edit is a powerful image editing foundation model achieving open-source state-of-the-art performance with precise instruction following, high-fidelity generation, superior identity consistency, and seamless multi-element fusion.

OmniPSD. Official code implementation of "OmniPSD: Layered PSD Generation with Diffusion Transformer"

Helios. Helios: Real Real-Time Long Video Generation Model

huobao-canvas. 火宝无限画布；支持文生图，图生图，图生视频，多模型切换。兼容 openai标准格式

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

ObjEmbed. (ICML 2026) Official repository of paper "ObjEmbed: Towards Universal Multimodal Object Embeddings"

Vec2Pix. HTML

Seed2.0. Jupyter Notebook

UniMRG. [ICML 2026] The official implementation of paper "Generation Enhances Understanding in Unified Multimodal Models via Multi-Representation Generation"

Kimi-K2.5. Moonshot's most powerful model

RAE. Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"

nanobot. Lightweight, open-source AI agent for your tools, chats, and workflows.

PrismLayers. PrismLayers: Open Data for High-Quality Multi-Layer Transparent Image Generative Models

Paper2Any. Turn paper/text/topic into editable research figures, technical route diagrams, and presentation slides.

awesome-ai-research-writing. Elevate your AI research writing, no more tedious polishing ✨

claude-code. Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

MIRROR. Python

DeepSeek-OCR-2. Visual Causal Flow

Edit-Banana. Edit Banana: A framework for converting statistical formats into editable.

Mind-Brush. Implement search image generation similar to Nano-banana-pro / Seedream / FLUX.

HADM. Code and dataset for "Detecting Human Artifacts from Text-to-Image Models"

Qwen-Image-Layered. Qwen-Image-Layered: Layered Decomposition for Inherent Editablity

InstantID. InstantID: Zero-shot Identity-Preserving Generation in Seconds 🔥

RISEBench. [NIPS 2025 DB Oral] Official Repository of paper: Envisioning Beyond the Pixels: Benchmarking Reasoning-Informed Visual Editing

score_sde_pytorch. PyTorch implementation for Score-Based Generative Modeling through Stochastic Differential Equations (ICLR 2021, Oral)

RectifiedFlow. Official Implementation of Rectified Flow (ICLR2023 Spotlight)

InternVLA-M1. InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policy

Open-PhyGDPO. PhyGDPO: Physics-Aware Groupwise Direct Preference Optimization for Physically Consistent Text-to-Video Generation (ECCV 2026)

vggt. [CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

DDPM. DDPM

Depth-Anything-3. Depth Anything 3

banana-slides. 一个基于nano banana pro🍌的原生AI PPT生成应用，迈向＂Vibe PPT＂; 支持上传任意模板图片，上传任意素材&智能解析，一句话/大纲/页面描述自动生成PPT，口头修改指定区域、一键导出可编辑ppt - An AI-native slides generator based on nano banana pro🍌

ml-unigen. UniGen: Enhanced Training & Test-Time Strategies for Unified Multimodal Understanding and Generation

X2DFD. Python

TwinFlow. [ICLR 2026] Taming large-scale few-step training with self-adversarial flows! 👏🏻

SegmentAnyRGBD. Segment Any RGBD

RynnVLA-002. RynnVLA-002: A Unified Vision-Language-Action and World Model

Seed-1.8. Jupyter Notebook

TRELLIS.2. Native and Compact Structured Latents for 3D Generation

LongBench-T2I. Draw ALL Your Imagine: A Holistic Benchmark and Agent Framework for Complex Instruction-based Image Generation

VideoAlign. [NeurIPS 2025] Improving Video Generation with Human Feedback

Ming. Ming - facilitating advanced multimodal understanding and generation capabilities built upon the Ling LLM.

LongCat-Image. Python

FLowMatching. Python

DiffSynth-Studio. Enjoy the magic of Diffusion models!

OmniAID. 「ICML 2026」Official PyTorch Code for "OmniAID: Decoupling Semantic and Artifacts for Universal AI-Generated Image Detection in the Wild".

DraCo. Offical Repository for Paper: DraCo: Draft as CoT for Text-to-Image Preview and Rare Concept Generation

RealGen. [ECCV 2026] RealGen: Photorealistic Text-to-Image Generation via Detector-Guided Rewards.

Adv-GRPO. [CVPR 2026] An official implementation of Adv-GRPO. The Image as Its Own Reward: Reinforcement Learning with Adversarial Reward for Image Generation.

UniSandBox. Does Understanding Inform Generation in Unified Multimodal Models? From Analysis to Path Forward

WriteHERE. An Open-Source AI Writing Project.

dmdr. [ECCV 2026] Official Code of "Distribution Matching Distillation Meets Reinforcement Learning"

Z-Image. Python

MajutsuCity. [CVPR' 26] MajutsuCity: Language-driven Aesthetic-adaptive City Generation with Controllable 3D Assets and Layouts

BLINK-Twice.

Paints-UNDO. Understand Human Behavior to Align True Needs

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

Emu3. Next-Token Prediction is All You Need

Emu3.5. Native Multimodal Models are World Learners

HivisionIDPhotos. ⚡️HivisionIDPhotos: a lightweight and efficient AI ID photos tools. 一个轻量级的AI证件照制作算法。

camel. 🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org

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

pico-banana-400k. Python

HunyuanWorld-Mirror. [ICML 2026] WorldMirror: Fast and Universal 3D reconstruction model for versatile tasks

UniGenBench. UniGenBench++: A Unified Semantic Evaluation Benchmark for Text-to-Image Generation

Edit-R1. Edit-R1: Reinforce Image Editing with Diffusion Negative-Aware Finetuning and MLLM Implicit Feedback

DeepSeek-OCR. Contexts Optical Compression

Earth-Agent. [ICLR 2026] The official implementation of the paper “Earth-Agent: Unlocking the Full Landscape of Earth Observation with Agents”

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

AI-Self-Training-DPO-SDXL. Unofficial implementation. Stable diffusion model trained by AI Feedback-Based Self-Training Direct Preference Optimization.

ULMEvalKit. ULMEvalKit: One-Stop Eval ToolKit for Image Generation

DA-2. Official implementation of "DA²: Depth Anything in Any Direction"

EasyR1. EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL

DeepSeek-V3.2-Exp. Python

HunyuanImage-3.0. HunyuanImage-3.0: A Powerful Native Multimodal Model for Image Generation

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

OmniDocBench. [CVPR 2025] A Comprehensive Benchmark for Document Parsing and Evaluation

verl. verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework

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

Logic-RL. Reproduce R1 Zero on Logic Puzzle

PromptEnhancer. [CVPR 2026] PromptEnhancer is a prompt-rewriting tool, refining prompts into clearer, structured versions for better image generation.

FastVideo. A unified inference and post-training framework for accelerated video generation.

reconstruction-alignment. [ICLR 2026] Official repo of paper "Reconstruction Alignment Improves Unified Multimodal Models". Unlocking the Massive Zero-shot Potential in Unified Multimodal Models through Self-supervised Learning.

VideoMMMU. Evaluating Knowledge Acquisition from Multi-Discipline Professional Videos

Textoon. Official repo for Textoon: Generating Vivid 2D Cartoon Characters from Text Descriptions

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

UAE. Official repository for the UAE paper, unified-GRPO, and unified-Bench

HunyuanImage-2.1. HunyuanImage-2.1: An Efficient Diffusion Model for High-Resolution (2K) Text-to-Image Generation​

Awesome-Nano-Banana-images. A curated collection of fun and creative examples generated with Nano Banana & Nano Banana Pro🍌, Gemini-2.5-flash-image based model. We also release Nano-consistent-150K openly to support the community's development of image generation and unified models(click to website to see our blog)

RecA-Page. HTML

ICEdit. [NeurIPS 2025] Image editing is worth a single LoRA! 0.1% training data for fantastic image editing! Surpasses GPT-4o in ID persistence~ MoE ckpt released! Only 4GB VRAM is enough to run!

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

MM-CelebA-HQ-Dataset. [CVPR 2021] Multi-Modal-CelebA-HQ: A Large-Scale Text-Driven Face Generation and Understanding Dataset