Awesome AI Agent Stack

Image, Video & Creative Generation

Diffusion, image and video generation tooling.

Image generation UIs

• AUTOMATIC1111/stable-diffusion-webui - Stable Diffusion web UI.
• Comfy-Org/ComfyUI - The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.
• lllyasviel/Fooocus - Focus on prompting and generating; Midjourney-style UI for SDXL.
• invoke-ai/InvokeAI - Creative engine for Stable Diffusion with a pro UI.
• lllyasviel/stable-diffusion-webui-forge - Stable Diffusion WebUI fork with faster inference and more features.
• vladmandic/sdnext - All-in-one WebUI for AI image/video creation and processing.
• Mooshieblob1/MooshieUI - Beginner-friendly front-end UI for ComfyUI.
• mcmonkeyprojects/SwarmUI - Modular Stable Diffusion web UI focused on power tools and speed.
• Acly/krita-ai-diffusion - Krita plugin for AI image generation, inpainting and outpainting.
• LykosAI/StabilityMatrix - Cross-platform package manager and launcher for Stable Diffusion UIs.
• MochiDiffusion/MochiDiffusion - Native macOS app for running Stable Diffusion locally.
• carson-katri/dream-textures - Stable Diffusion image and texture generation inside Blender.

Image models & inference

• huggingface/diffusers - State-of-the-art diffusion models for image, video and audio.
• black-forest-labs/flux - Official inference for FLUX image models.
• Stability-AI/generative-models - Stability AI generative model releases.
• apple-aiml-research/ml-stable-diffusion - Stable Diffusion with Core ML on Apple Silicon.
• ByteDance-Seed/Bagel - Open-source unified multimodal model for image generation and editing.
• Tencent-Hunyuan/HunyuanDiT - Multi-resolution diffusion transformer with fine language understanding.
• NVlabs/Sana - Efficient high-resolution text-to-image with linear diffusion transformers.
• QwenLM/Qwen-Image - Image generation foundation model with strong text rendering.
• stepfun-ai/Step1X-Edit - SOTA open-source instruction-based image editing model.
• leejet/stable-diffusion.cpp - Pure C/C++ inference for Stable Diffusion, Flux, Wan and other diffusion models.
• mflux-community/mflux - Apple MLX implementations of image and video generation models.

ComfyUI extensions

• Comfy-Org/ComfyUI-Manager - Extension manager and workflow hub for ComfyUI.
• ltdrdata/ComfyUI-Impact-Pack - Detection, detailer, and upscaling nodes for ComfyUI.
• ltdrdata/ComfyUI-Inspire-Pack - Versatile custom nodes extending ComfyUI workflows.
• Kosinkadink/ComfyUI-VideoHelperSuite - Video workflow nodes: loading, combining, and exporting.
• Kosinkadink/ComfyUI-AnimateDiff-Evolved - AnimateDiff video generation nodes for ComfyUI.
• Fannovel16/comfyui_controlnet_aux - ControlNet auxiliary preprocessors for ComfyUI.
• Kosinkadink/ComfyUI-Advanced-ControlNet - ControlNet scheduling and masking nodes for ComfyUI.
• kijai/ComfyUI-KJNodes - Handy utility and experimental nodes for ComfyUI.
• kijai/ComfyUI-WanVideoWrapper - Run WanVideo generation models inside ComfyUI.
• rgthree/rgthree-comfy - ComfyUI usability upgrades and workflow nodes.
• pythongosssss/ComfyUI-Custom-Scripts - UI enhancements and experiments for ComfyUI.
• yolain/ComfyUI-Easy-Use - Simplified nodes and presets making ComfyUI easier to use.
• 1038lab/ComfyUI-RMBG - ComfyUI nodes for AI background removal and segmentation.
• Gourieff/ComfyUI-ReActor - Fast face-swap extension node for ComfyUI.
• ssitu/ComfyUI_UltimateSDUpscale - Ultimate SD upscale tiling nodes for ComfyUI.
• MrForExample/ComfyUI-3D-Pack - Process 3D meshes and textures inside ComfyUI.
• city96/ComfyUI-GGUF - Run quantized GGUF diffusion models natively inside ComfyUI.
• ATH-MaaS/ComfyUI-Copilot - AI-powered ComfyUI assistant for workflow automation and guidance.
• SaladTechnologies/comfyui-api - Horizontally scalable ComfyUI API server with async outputs.
• Lightricks/ComfyUI-LTXVideo - ComfyUI nodes for LTX-Video text and image to video generation.
• nunchux-ai/ComfyUI-nunchaku - ComfyUI plugin for fast 4-bit quantized diffusion models via Nunchaku.
• numz/ComfyUI-SeedVR2_VideoUpscaler - ComfyUI nodes for SeedVR2 diffusion-based video and image upscaling.
• Comfy-Org/workflow_templates - Official ComfyUI template workflows.

Video generation

• calesthio/OpenMontage - World's first open-source, agentic video production system. 12 production pipelines, 100+ tools.
• showlab/Awesome-Video-Diffusion - A curated list of recent diffusion models for video generation, editing, and various other applications.
• hpcaitech/Open-Sora - Open video generation.
• Wan-Video/Wan2.1 - Alibaba's open text/image-to-video generation and video-editing model suite.
• Lightricks/LTX-Video - Lightricks' open real-time text-to-video generation model.
• mikehalleen/the-halleen-machine - Agentic system managing AI video projects via ComfyUI.
• genmoai/mochi - Open-source text-to-video generation models by Genmo.
• Tencent-Hunyuan/HunyuanVideo-1.5 - Lightweight open video generation model.
• zai-org/CogVideo - Open text-and-image-to-video generation (CogVideoX).
• Tencent-Hunyuan/HunyuanVideo - Large-scale open video generation model framework.
• SkyworkAI/SkyReels-V2 - Infinite-length AI film generation model.
• SkyworkAI/SkyReels-V3 - Multimodal video generation model.
• aigc-apps/VideoX-Fun - Flexible video generation at any resolution, from text or images.
• PKU-YuanGroup/Open-Sora-Plan - Open-source effort to reproduce Sora-class text-to-video.
• vita-epfl/Stable-Video-Infinity - Infinite-length video generation with error recycling.
• thu-ml/TurboDiffusion - 100-200x faster video diffusion inference.
• lllyasviel/FramePack - Practical long-video diffusion generation on consumer GPUs.
• LudwigKienle/ai-video-production-editor - Open-source AI video production editor with node pipelines.
• LinHao-city/StoryMind - LLM storyboard director that plans shots then generates AI films.
• harry0703/MoneyPrinterTurbo - Generates short videos from a topic with an automated LLM workflow.
• heygen-com/hyperframes - HTML-to-video rendering framework built for agents.
• meituan-longcat/LongCat-Video - Meituan's open video generation model.
• edenfunf/reelmimic - Agent crew that recreates the style of a video you show it.
• JayWebtech/autoshorts - Local desktop app that turns long recordings into vertical short clips.

Avatars & lip-sync

• KlingAIResearch/LivePortrait - Animate portraits into talking videos from a single photo.
• antgroup/echomimic - Lifelike audio-driven portrait animation with editable landmarks.
• jixiaozhong/Sonic - Audio-driven portrait animation with global audio perception.
• antgroup/echomimic_v3 - Unified multi-modal human animation model with 1.3B parameters.
• HumanAIGC/AnimateAnyone - Consistent controllable character animation from a single image.
• Rudrabha/Wav2Lip - Accurate lip-sync of any video to any audio.
• met4citizen/TalkingHead - JavaScript 3D talking heads with real-time lip-sync.
• chanjing-ai/Chanjing-Avatar - Open-source audio-driven 720p talking-avatar generation on Wan2.1.
• AISHIWEILAI/syncanimation - Real-time end-to-end audio-driven human animation framework.
• duixcom/Duix-Avatar - Open-source offline toolkit for AI avatars and digital human cloning.

Image editing, upscaling & restoration

• upscayl/upscayl - Free open-source AI image upscaler for desktop.
• chaiNNer-org/chaiNNer - Node-based image processing GUI for chaining upscaling tasks.
• k4yt3x/video2x - ML-based video super-resolution and frame interpolation.
• riponcm/nextgenUp - Free local AI image and video upscaler.
• Noah-1919/Fottly - Self-hosted AI image transformation and background removal service.
• danielgatis/rembg - Remove image backgrounds with AI, via CLI or library.
• sczhou/CodeFormer - Robust blind face restoration via codebook lookup transformers.
• ZhengPeng7/BiRefNet - High-resolution dichotomous image segmentation for matting.
• Akascape/Rembg-Fuse - Free open-source AI background removal plugin for DaVinci Resolve.
• pq-yang/MatAnyone - Stable video matting with consistent memory propagation.
• rizavelioglu/tryoffdiff - Virtual try-off: reconstruct garments from photos.
• tryonlabs/opentryon - Open APIs and models for virtual try-on fashion apps.
• nxnai/Voost - Unified diffusion transformer for virtual try-on and try-off.
• facefusion/facefusion - Industry-leading face swap and manipulation platform.
• hacksider/Deep-Live-Cam - Real-time face swap for live video from a single image.
• Nutlope/logocreator - Free open-source AI logo generator powered by Flux.
• Zeyi-Lin/HivisionIDPhotos - Lightweight AI ID photo maker.
• omnichar/OmniChar - AI character studio for consistent multi-model identity and LoRA training.
• zyddnys/manga-image-translator - OCR plus translation pipeline that rewrites text inside images and manga.

Training & fine-tuning

• Nerogar/OneTrainer - GUI and CLI toolkit for training and fine-tuning diffusion models including LoRA.
• kohya-ss/sd-scripts - Training scripts for DreamBooth, LoRA, and diffusion fine-tuning.
• bmaltais/kohya_ss - GUI for Kohya training scripts: DreamBooth, LoRA, and fine-tuning.
• ostris/ai-toolkit - Toolkit for fine-tuning diffusion models: LoRA and DreamBooth.
• cocktailpeanut/fluxgym - Dead-simple FLUX LoRA training UI with low VRAM support.
• UNfukashigi/SDXL-LoRA-Factory - GUI tool making SDXL LoRA training easy and intuitive.
• bghira/SimpleTuner - Fine-tuning toolkit for image, video and audio diffusion models.

Audio & music generation

• Stability-AI/stable-audio-tools - Conditional audio and music generation models.
• multimodal-art-projection/YuE - Frontier open music generation with lyrics and song structure.
• SonyResearch/Woosh - Sony's foundation model for sound effect generation.
• magenta/magenta-realtime - Open-weights live music generation model.
• xiaomi-research/controlfoley - Controllable video-to-audio generation model.

3D generation

• Tencent-Hunyuan/Hunyuan3D-2 - High-resolution 3D asset generation with diffusion models.
• microsoft/TRELLIS - Structured 3D latents for scalable and versatile 3D generation.
• Tencent-Hunyuan/HunyuanWorld-1.0 - Generate explorable 3D worlds from text or images.
• Tencent-Hunyuan/Hunyuan3D-2.1 - Image-to-3D assets with production-ready PBR materials.