600 AI models for images and video, open source
600 AI models for images and video, open source
Open Generative AI is an open source tool that lets you generate images and videos using over 600 models across 14 studios. No content filters or closed ecosystems. Great for creators and researchers exploring generative media.
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
Follow for daily open source drops 👇
[ar] 600 نموذج ذكاء اصطناعي للصور والفيديو، مفتوح المصدر
٦٠٠ نموذج ذكاء اصطناعي للصور والفيديوهات، مفتوح المصدر
Open Generative AI أداة مفتوحة المصدر تُمكّنك من إنشاء صور وفيديوهات باستخدام أكثر من ٦٠٠ نموذج من ١٤ استوديو. لا توجد فلاتر محتوى أو أنظمة مغلقة. مثالية للمبدعين والباحثين الذين يستكشفون الوسائط التوليدية.
anil-matcha/open-generative-ai
#الذكاء_الاصطناعي #مفتوح_المصدر #الذكاء_الاصطناعي_التوليدي #الذكاء_الاصطناعي_للفيديوهات
تابعنا للحصول على تحديثات يومية مفتوحة المصدر 👇
[bn] ছবি ও ভিডিওর জন্য ৬০০টি এআই মডেল, ওপেন সোর্স
ছবি এবং ভিডিওর জন্য ৬০০টি এআই মডেল, ওপেন সোর্স
ওপেন জেনারেটিভ এআই একটি ওপেন সোর্স টুল যা আপনাকে ১৪টি স্টুডিওর ৬০০টিরও বেশি মডেল ব্যবহার করে ছবি এবং ভিডিও তৈরি করতে দেয়। এতে কোনো কন্টেন্ট ফিল্টার বা সীমাবদ্ধ ইকোসিস্টেম নেই। জেনারেটিভ মিডিয়া নিয়ে কাজ করা নির্মাতা এবং গবেষকদের জন্য এটি দারুণ।
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
প্রতিদিন নতুন ওপেন সোর্স পেতে ফলো করুন 👇
[de-DE] 600 KI-Modelle für Bilder und Videos, Open Source
600 KI-Modelle für Bilder und Videos, Open Source
Open Generative AI ist ein Open-Source-Tool, mit dem Sie Bilder und Videos mithilfe von über 600 Modellen aus 14 Studios generieren können. Keine Inhaltsfilter oder geschlossenen Systeme. Ideal für Kreative und Forscher im Bereich generativer Medien.
anil-matcha/open-generative-ai
#KI #OpenSource #GenerativeKI #VideoKI
Folgen Sie uns für tägliche Open-Source-Neuheiten 👇
[es-US] 600 modelos de IA para imágenes y video, código abierto
600 modelos de IA para imágenes y vídeo, código abierto
Open Generative AI es una herramienta de código abierto que te permite generar imágenes y vídeos utilizando más de 600 modelos de 14 estudios. Sin filtros de contenido ni ecosistemas cerrados. Ideal para creadores e investigadores que exploran los medios generativos.
anil-matcha/open-generative-ai
#IA #CódigoAbierto #IAGenerativa #IAVídeo
Síguenos para recibir novedades diarias de código abierto 👇
[fr-FR] 600 modèles d'IA pour images et vidéos, en open source
600 modèles d'IA pour images et vidéos, open source
Open Generative AI est un outil open source permettant de générer des images et des vidéos grâce à plus de 600 modèles issus de 14 studios. Sans filtre de contenu ni écosystème fermé. Idéal pour les créateurs et les chercheurs explorant les médias génératifs.
anil-matcha/open-generative-ai
#IA #OpenSource #IAGénérative #IAVidéo
Suivez-nous pour découvrir chaque jour de nouveaux outils open source 👇
[hi] इमेज और वीडियो के लिए 600 AI मॉडल, ओपन सोर्स
इमेज और वीडियो के लिए 600 एआई मॉडल, ओपन सोर्स
ओपन जेनरेटिव एआई एक ओपन सोर्स टूल है जो आपको 14 स्टूडियो के 600 से अधिक मॉडलों का उपयोग करके इमेज और वीडियो बनाने की सुविधा देता है। इसमें कोई कंटेंट फ़िल्टर या बंद इकोसिस्टम नहीं है। जेनरेटिव मीडिया की खोज करने वाले रचनाकारों और शोधकर्ताओं के लिए यह बहुत उपयोगी है।
anil-matcha/open-generative-ai
#AI #ओपनसोर्स #जेनरेटिवएआई #वीडियोएआई
रोजाना ओपन सोर्स अपडेट के लिए फॉलो करें 👇
[id] 600 model AI untuk gambar dan video, sumber terbuka
600 Model AI untuk Gambar dan Video, Sumber Terbuka
Open Generative AI adalah alat sumber terbuka yang memungkinkan Anda menghasilkan gambar dan video menggunakan lebih dari 600 model di 14 studio. Tanpa filter konten atau ekosistem tertutup. Sangat cocok untuk kreator dan peneliti yang menjelajahi media generatif.
anil-matcha/open-generative-ai
#AI #SumberTerbuka #AIGeneratif #AIVideo
Ikuti untuk rilis sumber terbuka harian 👇
[it] 600 modelli di IA per immagini e video, open source
600 modelli di IA per immagini e video, open source
Open Generative AI è uno strumento open source che permette di generare immagini e video utilizzando oltre 600 modelli provenienti da 14 studi diversi. Nessun filtro sui contenuti o ecosistemi chiusi. Ideale per creatori e ricercatori che esplorano i media generativi.
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
Seguici per aggiornamenti quotidiani su progetti open source 👇
[iw] 600 מודלים של בינה מלאכותית לתמונות ווידאו, קוד פתוח
600 מודלים של בינה מלאכותית לתמונות ווידאו, קוד פתוח
בינה מלאכותית גנרטיבית פתוחה היא כלי קוד פתוח המאפשר לך ליצור תמונות וסרטונים באמצעות למעלה מ-600 מודלים ב-14 אולפנים. ללא מסנני תוכן או מערכות אקולוגיות סגורות. נהדר ליוצרים וחוקרים החוקרים מדיה גנרטיבית.
anil-matcha/open-generative-ai
#בינה מלאכותית #קודפתוח #בינה מלאכותיתגנרטיבית #וידאובינה מלאכותית
עקבו אחר פרסומים יומיים בקוד פתוח 👇
[ja] 画像・動画生成AIモデル600選、オープンソースで公開
画像と動画のための600種類以上のAIモデル、オープンソース
Open Generative AIは、14のスタジオから提供される600種類以上のモデルを使って画像や動画を生成できるオープンソースツールです。コンテンツフィルターや閉鎖的なエコシステムは一切ありません。ジェネレーティブメディアを探求するクリエイターや研究者に最適です。
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
毎日更新されるオープンソース情報をお見逃しなく👇
[ko] 이미지 및 영상 생성을 위한 600개의 오픈소스 AI 모델
이미지 및 비디오용 AI 모델 600개 이상, 오픈 소스
Open Generative AI는 14개 스튜디오에서 개발한 600개 이상의 모델을 사용하여 이미지와 비디오를 생성할 수 있는 오픈 소스 도구입니다. 콘텐츠 필터나 폐쇄적인 생태계가 없어 생성형 미디어를 탐구하는 크리에이터와 연구자에게 유용합니다.
anil-matcha/open-generative-ai
#AI #오픈소스 #생성형AI #비디오AI
매일 새로운 오픈 소스 소식을 받으려면 팔로우하세요 👇
[ml] ചിത്രങ്ങൾക്കും വീഡിയോകൾക്കുമായി 600 AI മോഡലുകൾ, ഓപ്പൺ സോഴ്സ്
ഇമേജുകൾക്കും വീഡിയോയ്ക്കുമായി 600 AI മോഡലുകൾ, ഓപ്പൺ സോഴ്സ്
14 സ്റ്റുഡിയോകളിലായി 600-ലധികം മോഡലുകൾ ഉപയോഗിച്ച് ചിത്രങ്ങളും വീഡിയോകളും സൃഷ്ടിക്കാൻ നിങ്ങളെ അനുവദിക്കുന്ന ഒരു ഓപ്പൺ സോഴ്സ് ഉപകരണമാണ് ഓപ്പൺ ജനറേറ്റീവ് AI. ഉള്ളടക്ക ഫിൽട്ടറുകളോ അടച്ച ആവാസവ്യവസ്ഥകളോ ഇല്ല. ജനറേറ്റീവ് മീഡിയ പര്യവേക്ഷണം ചെയ്യുന്ന സ്രഷ്ടാക്കൾക്കും ഗവേഷകർക്കും മികച്ചതാണ്.
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
പ്രതിദിന ഓപ്പൺ സോഴ്സ് ഡ്രോപ്പുകൾക്കായി പിന്തുടരുക 👇
[nl-NL] 600 AI-modellen voor beeld en video, open source
600 AI-modellen voor afbeeldingen en video, open source
Open Generative AI is een open source tool waarmee je afbeeldingen en video's kunt genereren met behulp van meer dan 600 modellen uit 14 studio's. Geen contentfilters of gesloten ecosystemen. Ideaal voor makers en onderzoekers die generatieve media verkennen.
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
Volg ons voor dagelijkse open source updates 👇
[pa] ਚਿੱਤਰਾਂ ਅਤੇ ਵੀਡੀਓ ਲਈ 600 ਏਆਈ (AI) ਮਾਡਲ, ਓਪਨ ਸੋਰਸ
ਤਸਵੀਰਾਂ ਅਤੇ ਵੀਡੀਓ ਲਈ 600 AI ਮਾਡਲ, ਓਪਨ ਸੋਰਸ
ਓਪਨ ਜਨਰੇਟਿਵ AI ਇੱਕ ਓਪਨ ਸੋਰਸ ਟੂਲ ਹੈ ਜੋ ਤੁਹਾਨੂੰ 14 ਸਟੂਡੀਓ ਵਿੱਚ 600 ਤੋਂ ਵੱਧ ਮਾਡਲਾਂ ਦੀ ਵਰਤੋਂ ਕਰਕੇ ਤਸਵੀਰਾਂ ਅਤੇ ਵੀਡੀਓ ਤਿਆਰ ਕਰਨ ਦਿੰਦਾ ਹੈ। ਕੋਈ ਸਮੱਗਰੀ ਫਿਲਟਰ ਜਾਂ ਬੰਦ ਈਕੋਸਿਸਟਮ ਨਹੀਂ। ਜਨਰੇਟਿਵ ਮੀਡੀਆ ਦੀ ਪੜਚੋਲ ਕਰਨ ਵਾਲੇ ਸਿਰਜਣਹਾਰਾਂ ਅਤੇ ਖੋਜਕਰਤਾਵਾਂ ਲਈ ਵਧੀਆ।
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
ਰੋਜ਼ਾਨਾ ਓਪਨ ਸੋਰਸ ਡ੍ਰੌਪਸ ਲਈ ਫਾਲੋ ਕਰੋ 👇
[pl] 600 modeli AI do generowania obrazów i wideo, open source
600 modeli AI do obrazów i wideo, open source
Open Generative AI to narzędzie open source, które pozwala generować obrazy i filmy przy użyciu ponad 600 modeli w 14 studiach. Bez filtrów treści i zamkniętych ekosystemów. Idealne dla twórców i badaczy eksplorujących media generatywne.
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
Obserwuj, aby otrzymywać codzienne informacje o open source 👇
[pt-BR] 600 modelos de IA para imagens e vídeo, código aberto
600 modelos de IA para imagens e vídeos, código aberto
O Open Generative AI é uma ferramenta de código aberto que permite gerar imagens e vídeos usando mais de 600 modelos de 14 estúdios diferentes. Sem filtros de conteúdo ou ecossistemas fechados. Ideal para criadores e pesquisadores que exploram mídia generativa.
anil-matcha/open-generative-ai
#IA #CódigoAberto #IAGenerativa #IAVídeo
Siga para receber novidades diárias sobre código aberto 👇
[ru] 600 нейросетей для генерации изображений и видео с открытым исходным кодом
600 моделей ИИ для изображений и видео, открытый исходный код
Open Generative AI — это инструмент с открытым исходным кодом, позволяющий создавать изображения и видео, используя более 600 моделей из 14 студий. Никаких контентных фильтров или закрытых экосистем. Отлично подходит для создателей и исследователей, изучающих генеративные медиа.
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
Подписывайтесь, чтобы получать ежедневные обновления с открытым исходным кодом 👇
[ta] 600 ஏஐ மாதிரிகள் மூலம் படங்கள் மற்றும் காணொலிகள், திறந்த மூல மென்பொருள்
படங்கள் மற்றும் வீடியோவிற்கான 600 AI மாடல்கள், ஓப்பன் சோர்ஸ்
ஓப்பன் ஜெனரேட்டிவ் AI என்பது ஒரு ஓப்பன் சோர்ஸ் கருவியாகும். இது 14 ஸ்டுடியோக்களில் உள்ள 600-க்கும் மேற்பட்ட மாடல்களைப் பயன்படுத்தி, படங்கள் மற்றும் வீடியோக்களை உருவாக்க உங்களை அனுமதிக்கிறது. இதில் உள்ளடக்க வடிகட்டிகளோ அல்லது மூடிய சூழல் அமைப்புகளோ இல்லை. ஜெனரேட்டிவ் மீடியாவை ஆராயும் படைப்பாளர்களுக்கும் ஆராய்ச்சியாளர்களுக்கும் இது மிகவும் சிறந்தது.
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
தினசரி ஓப்பன் சோர்ஸ் வெளியீடுகளுக்குப் பின்தொடரவும் 👇
[te] చిత్రాలు మరియు వీడియోల కోసం 600 AI మోడల్స్, ఓపెన్ సోర్స్
చిత్రాలు మరియు వీడియోల కోసం 600 AI మోడల్స్, ఓపెన్ సోర్స్
ఓపెన్ జనరేటివ్ AI అనేది ఒక ఓపెన్ సోర్స్ సాధనం. ఇది 14 స్టూడియోల నుండి 600కు పైగా మోడల్స్ను ఉపయోగించి చిత్రాలు మరియు వీడియోలను రూపొందించడానికి మిమ్మల్ని అనుమతిస్తుంది. కంటెంట్ ఫిల్టర్లు లేదా క్లోజ్డ్ ఎకోసిస్టమ్స్ ఉండవు. జనరేటివ్ మీడియాను అన్వేషించే సృష్టికర్తలు మరియు పరిశోధకులకు ఇది చాలా గొప్పది.
anil-matcha/open-generative-ai
#AI #Opensource #GenerativeAI #VideoAI
ప్రతిరోజూ కొత్త ఓపెన్ సోర్స్ అప్డేట్ల కోసం ఫాలో అవ్వండి 👇
[uk] 600 ШІ-моделей для зображень та відео, з відкритим кодом
600 моделей ШІ для зображень та відео, з відкритим вихідним кодом
Open Generative AI – це інструмент з відкритим вихідним кодом, який дозволяє створювати зображення та відео, використовуючи понад 600 моделей у 14 студіях. Без фільтрів контенту чи закритих екосистем. Чудово підходить для творців та дослідників, які досліджують генеративні медіа.
anil-matcha/open-generative-ai
#ШІ #ВідкритийВихіднийКод #ГенеративнийШІ #ВідеоШІ
Підпишіться на щоденні оновлення з відкритим вихідним кодом 👇
147 views
Open Generative AI GitHub Repo: A Self-Hosting Teardown (2026)
TL;DR: If you search "open generative ai github" in 2026, the repo you land on is Anil-matcha/Open-Generative-AI: an MIT-licensed Next.js and Electron studio for image, video, and lip sync generation, at roughly 29.7k stars and 5.4k forks as of October 6. The UI is open source and self-hostable. Most of the models behind it are not. Cloud generation goes through the Muapi.ai API with your own key, and truly local inference only exists in the desktop app, through two engines with real hardware requirements.
Some context on why I care. I build media features into client apps, mostly backend work: queues, webhooks, storage. When a client wants text-to-video in production, I usually call a hosted API such as AI Video API rather than run video models on our own GPUs. Open Generative AI keeps coming up as the "free, self-hosted" alternative, so I read through the repo, the package scripts, and the release assets to see what you actually get.
1. What the Open Generative AI GitHub repo is
The dev.to article that currently ranks for this search is a January 2025 list of three image projects: CompVis Stable Diffusion, DALL-E Mini, and StyleGAN3. Those are model repos. Open Generative AI is a different kind of thing: a front end that puts hundreds of hosted models behind one interface.
From the README and package.json:
- Stack: Next.js 14 (App Router), React 18, Tailwind CSS v3, npm workspaces. A shared packages/studio component library holds the studios, and packages/studio/src/models.js is the single list of model definitions.
- Studios: Image, Video, Audio, Lip Sync, Cinema (camera, lens, focal length, and aperture presets turned into prompt modifiers), Workflow (node-based pipelines), Agent, Design Agent, Marketing, AI Clipping, Body Swap, and a few more.
- Distribution: a web build (npm run dev), an Electron desktop build, and a hosted copy on muapi.ai.
- License: MIT.
The same packages/studio library powers the hosted version on muapi.ai, which explains why the model list moves so fast. Model updates land in one file and ship to both.
2. What's open and what isn't
This is the part the repo description ("Self-hosted, MIT licensed") glosses over.
| Layer | Where it runs | What you need |
| --- | --- | --- |
| UI (all studios) | Your machine or server | Node.js 18+, or the desktop installer |
| Cloud models (Flux, Kling, Veo, Sora, Seedance, Midjourney...) | Muapi's API | A Muapi access key, billed by Muapi |
| sd.cpp local engine | Your machine, desktop app only | CPU works; Metal on Apple Silicon, CUDA/Vulkan/ROCm elsewhere |
| Wan2GP local engine | Your GPU box, desktop app only | Your own Wan2GP install on a CUDA or ROCm GPU |
| Hosted web version | muapi.ai | A free account; always uses cloud APIs |
The API flow is documented in the README. The app submits a job with POST /api/v1/{model-endpoint}, then polls GET /api/v1/predictions/{request_id}/result until the status is completed, authenticating with an x-api-key header. Your key sits in browser localStorage and, per the README, is only sent to Muapi.
So "free" means the code costs nothing. Generations on the big commercial models cost whatever Muapi charges for them. That's a perfectly reasonable design, a bring-your-own-key client, but you should budget for it before you tell a stakeholder it's free.
3. Running it from source
The README is clear that most people should grab a prebuilt installer. If you want to hack on it, this is the path, and I checked each script name against package.json:
# submodules are required for the workflow and agent packages
git clone --recurse-submodules https://github.com/Anil-matcha/Open-Generative-AI.git
cd Open-Generative-AI
# "setup" = git submodule update + npm install + build all workspace packages
# plain `npm install` is not enough
npm run setup
# then ONE of these
npm run electron:dev # desktop app (Vite build, then Electron)
npm run dev # web version on http://localhost:3000
# production web build
npm run build && npm run start
If Next.js complains it can't find a pages directory, you're either not in the repo root or you cloned without submodules. Re-run npm run setup if packages/Vibe-Workflow or packages/agents are empty.
4. Local inference: two engines, very different costs
Local generation is desktop-only. The hosted and web builds always call cloud APIs.
sd.cpp (bundled). Built on stable-diffusion.cpp. You install it with one click under Settings, then Local Models. It handles image models only: Dreamshaper 8, Realistic Vision 5.1, and Anything v5 (SD 1.5, about 2.1 GB each), SDXL Base 1.0 (6.9 GB), and Z-Image Turbo and Base, which also need a 2.4 GB Qwen3-4B text encoder and a 335 MB FLUX VAE. The README warns that Z-Image is known to hang a base 8 GB M-series Mac and recommends 16 GB of RAM. On 8 GB, stick to SD 1.5.
Weights default to Electron's app-data folder (~/Library/Application Support/open-generative-ai/local-ai on macOS, ~/.config/open-generative-ai/local-ai on Linux). If you'd rather keep multi-GB weights on another drive, set OPEN_GENERATIVE_AI_LOCAL_AI_DIR before launch.
Wan2GP (bring your own). This is where video lives. The app doesn't bundle Python or weights. You install Wan2GP yourself on a CUDA or ROCm machine, and the app either runs wgp.py from a local folder or talks to Wan2GP's MCP server on another box:
python wgp.py --mcp --mcp-api-version 1 --mcp-transport streamable-http \
--mcp-host 0.0.0.0 --mcp-port 7866
You paste http://<host>:7866/mcp into the settings panel. The Gradio UI on port 7860 won't work for this. Only bind to 0.0.0.0 on a network you trust.
The remote mode is the clever part. Wan2GP has no Apple Silicon path, so a Mac user can keep the desktop app and push inference to a Linux GPU box or a rented instance. Supported models include Wan 2.1 1.3B (480p, about 8 GB VRAM, the lightest option), Wan 2.2 14B T2V/I2V, Hunyuan Video 1.5, LTX-2 Distilled (720p), plus Flux.1 Dev and Qwen Image for stills.
5. Gotchas I'd check before relying on it
- The numbers in the README disagree with each other. The tagline says 400+ models, a features bullet says 200+, and the category table says 420+ "verified against models.js." Lip sync is "9 dedicated models" in one section and 15 in the category table. Count models.js yourself if the number matters to you.
- The download table is stale. It still links 1.0.9 installers, while the latest GitHub release is v2.0.0 (May 23, 2026), with .dmg builds for arm64 and Intel, a Windows .exe, an .AppImage, and a .deb. Go to the Releases page directly.
- The binaries are unsigned. On macOS you'll need xattr -cr "/Applications/Open Generative AI.app" or the "Open Anyway" button; on Windows, SmartScreen's "Run anyway."
- Ubuntu 24.04 AppArmor. The AppImage can die silently because of apparmor_restrict_unprivileged_userns. The .deb ships an AppArmor profile, so install that instead of flipping the sysctl.
- "No content filters" describes the UI. Your generations still run on Muapi and on the upstream model providers, so read their terms before you build a product on that promise.
- The README sells white-label hosting. MuAPI's white-label plan starts at $49/mo. It's optional, but expect upsell links throughout the docs.
6. How it compares to the older GitHub picks
| Repo | Stars (Oct 2026) | Last push | What it is |
| --- | --- | --- | --- |
| Anil-matcha/Open-Generative-AI | 29.7k | Oct 2026 | Multi-model studio UI, BYO key plus local engines |
| CompVis/stable-diffusion | 73.5k | Jun 2024 | Original SD research code |
| borisdayma/dalle-mini | 14.7k | Nov 2023 | DALL-E Mini model code |
| NVlabs/stylegan3 | 6.9k | Sep 2023 | GAN research code, NVIDIA GPU |
| deepbeepmeep/Wan2GP | 10.1k | Oct 2026 | Local video/image runtime for modest GPUs |
| leejet/stable-diffusion.cpp | 7.5k | Oct 2026 | C++ diffusion inference, powers the sd.cpp engine |
The 2025 list is still useful as history. If you want to generate something this week, the three bottom rows are where the activity is, and the top row ties them together behind a UI.
FAQ
Is Open Generative AI on GitHub really free?
The code is MIT licensed and free. Cloud generations go through Muapi with your own key, which is billed. Local generation through sd.cpp or Wan2GP costs only your hardware.
Can I run Open Generative AI fully offline?
Only in the desktop app, using sd.cpp for images or a Wan2GP install for video. The web build and the hosted version always use cloud APIs.
What GPU do I need for local video?
Wan2GP needs a CUDA or ROCm GPU. Wan 2.1 1.3B at 480p is listed at about 8 GB VRAM. Larger models such as Wan 2.2 14B are noted as slow on consumer cards.
Is Open Generative AI the same as Stable Diffusion?
No. Stable Diffusion is a model. Open Generative AI is a studio that can run SD 1.5 and SDXL locally and also calls hundreds of hosted models.
Where I landed
For prototyping and comparing models side by side, Open Generative AI is the most complete open-source front end I've looked at, and the Wan2GP-over-MCP setup is worth stealing even if you never use the UI. For a production feature where a user clicks "generate" and expects a video back through a webhook, I still prefer a narrower hosted service. For that job I reach for AI Video API, which handles text-to-video and image-to-video with callback webhooks, so my backend stays a queue and a handler. The repo is at Anil-matcha/Open-Generative-AI if you want to read models.js yourself.
5 reactions · 4 comments
Mentions
Open Higgsfield AI: 400+ AI Image & Video Models in One Free Studio #Shorts Open Higgsfield AI (now Open Generative AI on GitHub) is a free, open-source AI image and video studio: Flux, Midjourney, Kling, Sora, Veo and more, with 420+ models across 8 categories in the README's model table. Lip Sync Studio, Cinema Studio camera controls, up to 14 reference images, Workflow Studio pipelines, and local models in the desktop app. MIT licensed, with 31,500+ GitHub stars. Full video: youtu.be Source: Anil-matcha/Open-Generative-AI Not financial advice. #AI #OpenSource #GenerativeAI #AIVideo #AIImage #GitHub
600 AI models for images and video, open source 600 AI models for images and video, open source Open Generative AI is an open source tool that lets you generate images and videos using over 600 models across 14 studios. No content filters or closed ecosystems. Great for creators and researchers exploring generative media. anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI Follow for daily open source drops 👇 [ar] 600 نموذج ذكاء اصطناعي للصور والفيديو، مفتوح المصدر ٦٠٠ نموذج ذكاء اصطناعي للصور والفيديوهات، مفتوح المصدر Open Generative AI أداة مفتوحة المصدر تُمكّنك من إنشاء صور وفيديوهات باستخدام أكثر من ٦٠٠ نموذج من ١٤ استوديو. لا توجد فلاتر محتوى أو أنظمة مغلقة. مثالية للمبدعين والباحثين الذين يستكشفون الوسائط التوليدية. anil-matcha/open-generative-ai #الذكاء_الاصطناعي #مفتوح_المصدر #الذكاء_الاصطناعي_التوليدي #الذكاء_الاصطناعي_للفيديوهات تابعنا للحصول على تحديثات يومية مفتوحة المصدر 👇 [bn] ছবি ও ভিডিওর জন্য ৬০০টি এআই মডেল, ওপেন সোর্স ছবি এবং ভিডিওর জন্য ৬০০টি এআই মডেল, ওপেন সোর্স ওপেন জেনারেটিভ এআই একটি ওপেন সোর্স টুল যা আপনাকে ১৪টি স্টুডিওর ৬০০টিরও বেশি মডেল ব্যবহার করে ছবি এবং ভিডিও তৈরি করতে দেয়। এতে কোনো কন্টেন্ট ফিল্টার বা সীমাবদ্ধ ইকোসিস্টেম নেই। জেনারেটিভ মিডিয়া নিয়ে কাজ করা নির্মাতা এবং গবেষকদের জন্য এটি দারুণ। anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI প্রতিদিন নতুন ওপেন সোর্স পেতে ফলো করুন 👇 [de-DE] 600 KI-Modelle für Bilder und Videos, Open Source 600 KI-Modelle für Bilder und Videos, Open Source Open Generative AI ist ein Open-Source-Tool, mit dem Sie Bilder und Videos mithilfe von über 600 Modellen aus 14 Studios generieren können. Keine Inhaltsfilter oder geschlossenen Systeme. Ideal für Kreative und Forscher im Bereich generativer Medien. anil-matcha/open-generative-ai #KI #OpenSource #GenerativeKI #VideoKI Folgen Sie uns für tägliche Open-Source-Neuheiten 👇 [es-US] 600 modelos de IA para imágenes y video, código abierto 600 modelos de IA para imágenes y vídeo, código abierto Open Generative AI es una herramienta de código abierto que te permite generar imágenes y vídeos utilizando más de 600 modelos de 14 estudios. Sin filtros de contenido ni ecosistemas cerrados. Ideal para creadores e investigadores que exploran los medios generativos. anil-matcha/open-generative-ai #IA #CódigoAbierto #IAGenerativa #IAVídeo Síguenos para recibir novedades diarias de código abierto 👇 [fr-FR] 600 modèles d'IA pour images et vidéos, en open source 600 modèles d'IA pour images et vidéos, open source Open Generative AI est un outil open source permettant de générer des images et des vidéos grâce à plus de 600 modèles issus de 14 studios. Sans filtre de contenu ni écosystème fermé. Idéal pour les créateurs et les chercheurs explorant les médias génératifs. anil-matcha/open-generative-ai #IA #OpenSource #IAGénérative #IAVidéo Suivez-nous pour découvrir chaque jour de nouveaux outils open source 👇 [hi] इमेज और वीडियो के लिए 600 AI मॉडल, ओपन सोर्स इमेज और वीडियो के लिए 600 एआई मॉडल, ओपन सोर्स ओपन जेनरेटिव एआई एक ओपन सोर्स टूल है जो आपको 14 स्टूडियो के 600 से अधिक मॉडलों का उपयोग करके इमेज और वीडियो बनाने की सुविधा देता है। इसमें कोई कंटेंट फ़िल्टर या बंद इकोसिस्टम नहीं है। जेनरेटिव मीडिया की खोज करने वाले रचनाकारों और शोधकर्ताओं के लिए यह बहुत उपयोगी है। anil-matcha/open-generative-ai #AI #ओपनसोर्स #जेनरेटिवएआई #वीडियोएआई रोजाना ओपन सोर्स अपडेट के लिए फॉलो करें 👇 [id] 600 model AI untuk gambar dan video, sumber terbuka 600 Model AI untuk Gambar dan Video, Sumber Terbuka Open Generative AI adalah alat sumber terbuka yang memungkinkan Anda menghasilkan gambar dan video menggunakan lebih dari 600 model di 14 studio. Tanpa filter konten atau ekosistem tertutup. Sangat cocok untuk kreator dan peneliti yang menjelajahi media generatif. anil-matcha/open-generative-ai #AI #SumberTerbuka #AIGeneratif #AIVideo Ikuti untuk rilis sumber terbuka harian 👇 [it] 600 modelli di IA per immagini e video, open source 600 modelli di IA per immagini e video, open source Open Generative AI è uno strumento open source che permette di generare immagini e video utilizzando oltre 600 modelli provenienti da 14 studi diversi. Nessun filtro sui contenuti o ecosistemi chiusi. Ideale per creatori e ricercatori che esplorano i media generativi. anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI Seguici per aggiornamenti quotidiani su progetti open source 👇 [iw] 600 מודלים של בינה מלאכותית לתמונות ווידאו, קוד פתוח 600 מודלים של בינה מלאכותית לתמונות ווידאו, קוד פתוח בינה מלאכותית גנרטיבית פתוחה היא כלי קוד פתוח המאפשר לך ליצור תמונות וסרטונים באמצעות למעלה מ-600 מודלים ב-14 אולפנים. ללא מסנני תוכן או מערכות אקולוגיות סגורות. נהדר ליוצרים וחוקרים החוקרים מדיה גנרטיבית. anil-matcha/open-generative-ai #בינה מלאכותית #קודפתוח #בינה מלאכותיתגנרטיבית #וידאובינה מלאכותית עקבו אחר פרסומים יומיים בקוד פתוח 👇 [ja] 画像・動画生成AIモデル600選、オープンソースで公開 画像と動画のための600種類以上のAIモデル、オープンソース Open Generative AIは、14のスタジオから提供される600種類以上のモデルを使って画像や動画を生成できるオープンソースツールです。コンテンツフィルターや閉鎖的なエコシステムは一切ありません。ジェネレーティブメディアを探求するクリエイターや研究者に最適です。 anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI 毎日更新されるオープンソース情報をお見逃しなく👇 [ko] 이미지 및 영상 생성을 위한 600개의 오픈소스 AI 모델 이미지 및 비디오용 AI 모델 600개 이상, 오픈 소스 Open Generative AI는 14개 스튜디오에서 개발한 600개 이상의 모델을 사용하여 이미지와 비디오를 생성할 수 있는 오픈 소스 도구입니다. 콘텐츠 필터나 폐쇄적인 생태계가 없어 생성형 미디어를 탐구하는 크리에이터와 연구자에게 유용합니다. anil-matcha/open-generative-ai #AI #오픈소스 #생성형AI #비디오AI 매일 새로운 오픈 소스 소식을 받으려면 팔로우하세요 👇 [ml] ചിത്രങ്ങൾക്കും വീഡിയോകൾക്കുമായി 600 AI മോഡലുകൾ, ഓപ്പൺ സോഴ്സ് ഇമേജുകൾക്കും വീഡിയോയ്ക്കുമായി 600 AI മോഡലുകൾ, ഓപ്പൺ സോഴ്സ് 14 സ്റ്റുഡിയോകളിലായി 600-ലധികം മോഡലുകൾ ഉപയോഗിച്ച് ചിത്രങ്ങളും വീഡിയോകളും സൃഷ്ടിക്കാൻ നിങ്ങളെ അനുവദിക്കുന്ന ഒരു ഓപ്പൺ സോഴ്സ് ഉപകരണമാണ് ഓപ്പൺ ജനറേറ്റീവ് AI. ഉള്ളടക്ക ഫിൽട്ടറുകളോ അടച്ച ആവാസവ്യവസ്ഥകളോ ഇല്ല. ജനറേറ്റീവ് മീഡിയ പര്യവേക്ഷണം ചെയ്യുന്ന സ്രഷ്ടാക്കൾക്കും ഗവേഷകർക്കും മികച്ചതാണ്. anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI പ്രതിദിന ഓപ്പൺ സോഴ്സ് ഡ്രോപ്പുകൾക്കായി പിന്തുടരുക 👇 [nl-NL] 600 AI-modellen voor beeld en video, open source 600 AI-modellen voor afbeeldingen en video, open source Open Generative AI is een open source tool waarmee je afbeeldingen en video's kunt genereren met behulp van meer dan 600 modellen uit 14 studio's. Geen contentfilters of gesloten ecosystemen. Ideaal voor makers en onderzoekers die generatieve media verkennen. anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI Volg ons voor dagelijkse open source updates 👇 [pa] ਚਿੱਤਰਾਂ ਅਤੇ ਵੀਡੀਓ ਲਈ 600 ਏਆਈ (AI) ਮਾਡਲ, ਓਪਨ ਸੋਰਸ ਤਸਵੀਰਾਂ ਅਤੇ ਵੀਡੀਓ ਲਈ 600 AI ਮਾਡਲ, ਓਪਨ ਸੋਰਸ ਓਪਨ ਜਨਰੇਟਿਵ AI ਇੱਕ ਓਪਨ ਸੋਰਸ ਟੂਲ ਹੈ ਜੋ ਤੁਹਾਨੂੰ 14 ਸਟੂਡੀਓ ਵਿੱਚ 600 ਤੋਂ ਵੱਧ ਮਾਡਲਾਂ ਦੀ ਵਰਤੋਂ ਕਰਕੇ ਤਸਵੀਰਾਂ ਅਤੇ ਵੀਡੀਓ ਤਿਆਰ ਕਰਨ ਦਿੰਦਾ ਹੈ। ਕੋਈ ਸਮੱਗਰੀ ਫਿਲਟਰ ਜਾਂ ਬੰਦ ਈਕੋਸਿਸਟਮ ਨਹੀਂ। ਜਨਰੇਟਿਵ ਮੀਡੀਆ ਦੀ ਪੜਚੋਲ ਕਰਨ ਵਾਲੇ ਸਿਰਜਣਹਾਰਾਂ ਅਤੇ ਖੋਜਕਰਤਾਵਾਂ ਲਈ ਵਧੀਆ। anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI ਰੋਜ਼ਾਨਾ ਓਪਨ ਸੋਰਸ ਡ੍ਰੌਪਸ ਲਈ ਫਾਲੋ ਕਰੋ 👇 [pl] 600 modeli AI do generowania obrazów i wideo, open source 600 modeli AI do obrazów i wideo, open source Open Generative AI to narzędzie open source, które pozwala generować obrazy i filmy przy użyciu ponad 600 modeli w 14 studiach. Bez filtrów treści i zamkniętych ekosystemów. Idealne dla twórców i badaczy eksplorujących media generatywne. anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI Obserwuj, aby otrzymywać codzienne informacje o open source 👇 [pt-BR] 600 modelos de IA para imagens e vídeo, código aberto 600 modelos de IA para imagens e vídeos, código aberto O Open Generative AI é uma ferramenta de código aberto que permite gerar imagens e vídeos usando mais de 600 modelos de 14 estúdios diferentes. Sem filtros de conteúdo ou ecossistemas fechados. Ideal para criadores e pesquisadores que exploram mídia generativa. anil-matcha/open-generative-ai #IA #CódigoAberto #IAGenerativa #IAVídeo Siga para receber novidades diárias sobre código aberto 👇 [ru] 600 нейросетей для генерации изображений и видео с открытым исходным кодом 600 моделей ИИ для изображений и видео, открытый исходный код Open Generative AI — это инструмент с открытым исходным кодом, позволяющий создавать изображения и видео, используя более 600 моделей из 14 студий. Никаких контентных фильтров или закрытых экосистем. Отлично подходит для создателей и исследователей, изучающих генеративные медиа. anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI Подписывайтесь, чтобы получать ежедневные обновления с открытым исходным кодом 👇 [ta] 600 ஏஐ மாதிரிகள் மூலம் படங்கள் மற்றும் காணொலிகள், திறந்த மூல மென்பொருள் படங்கள் மற்றும் வீடியோவிற்கான 600 AI மாடல்கள், ஓப்பன் சோர்ஸ் ஓப்பன் ஜெனரேட்டிவ் AI என்பது ஒரு ஓப்பன் சோர்ஸ் கருவியாகும். இது 14 ஸ்டுடியோக்களில் உள்ள 600-க்கும் மேற்பட்ட மாடல்களைப் பயன்படுத்தி, படங்கள் மற்றும் வீடியோக்களை உருவாக்க உங்களை அனுமதிக்கிறது. இதில் உள்ளடக்க வடிகட்டிகளோ அல்லது மூடிய சூழல் அமைப்புகளோ இல்லை. ஜெனரேட்டிவ் மீடியாவை ஆராயும் படைப்பாளர்களுக்கும் ஆராய்ச்சியாளர்களுக்கும் இது மிகவும் சிறந்தது. anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI தினசரி ஓப்பன் சோர்ஸ் வெளியீடுகளுக்குப் பின்தொடரவும் 👇 [te] చిత్రాలు మరియు వీడియోల కోసం 600 AI మోడల్స్, ఓపెన్ సోర్స్ చిత్రాలు మరియు వీడియోల కోసం 600 AI మోడల్స్, ఓపెన్ సోర్స్ ఓపెన్ జనరేటివ్ AI అనేది ఒక ఓపెన్ సోర్స్ సాధనం. ఇది 14 స్టూడియోల నుండి 600కు పైగా మోడల్స్ను ఉపయోగించి చిత్రాలు మరియు వీడియోలను రూపొందించడానికి మిమ్మల్ని అనుమతిస్తుంది. కంటెంట్ ఫిల్టర్లు లేదా క్లోజ్డ్ ఎకోసిస్టమ్స్ ఉండవు. జనరేటివ్ మీడియాను అన్వేషించే సృష్టికర్తలు మరియు పరిశోధకులకు ఇది చాలా గొప్పది. anil-matcha/open-generative-ai #AI #Opensource #GenerativeAI #VideoAI ప్రతిరోజూ కొత్త ఓపెన్ సోర్స్ అప్డేట్ల కోసం ఫాలో అవ్వండి 👇 [uk] 600 ШІ-моделей для зображень та відео, з відкритим кодом 600 моделей ШІ для зображень та відео, з відкритим вихідним кодом Open Generative AI – це інструмент з відкритим вихідним кодом, який дозволяє створювати зображення та відео, використовуючи понад 600 моделей у 14 студіях. Без фільтрів контенту чи закритих екосистем. Чудово підходить для творців та дослідників, які досліджують генеративні медіа. anil-matcha/open-generative-ai #ШІ #ВідкритийВихіднийКод #ГенеративнийШІ #ВідеоШІ Підпишіться на щоденні оновлення з відкритим вихідним кодом 👇
10/10 — Open Generative AI An open-source creative studio for AI images and videos. Generate visuals, create lip-sync content, and build custom creative workflows. anil-matcha/open-generative-ai t.co
Project link :- Anil-matcha/Open-Generative-AI Give a star if you found it useful
Companies are raising 10M$ funds for AI Video flows But I am open-sourcing the same for free Introducing Open-Generative-AI (30k+ stars already) 600+ models, no subscription fee, no lock-in, no restrictions Repo -> Anil-matcha/Open-Generative-AI
Did the same with Open Generative AI which now has 30k+ stars Open-source alternative to all paid tools like Invideo, Runwayml, OpenArt and so on Killed all subscriptions and made it free Repo link :- Anil-matcha/Open-Generative-AI
@kingcodru Check Anil-matcha/Open-Generative-AI it's already a popular repo which does the same
Open Generative AI GitHub Repo: A Self-Hosting Teardown (2026) TL;DR: If you search "open generative ai github" in 2026, the repo you land on is Anil-matcha/Open-Generative-AI: an MIT-licensed Next.js and Electron studio for image, video, and lip sync generation, at roughly 29.7k stars and 5.4k forks as of October 6. The UI is open source and self-hostable. Most of the models behind it are not. Cloud generation goes through the Muapi.ai API with your own key, and truly local inference only exists in the desktop app, through two engines with real hardware requirements. Some context on why I care. I build media features into client apps, mostly backend work: queues, webhooks, storage. When a client wants text-to-video in production, I usually call a hosted API such as AI Video API rather than run video models on our own GPUs. Open Generative AI keeps coming up as the "free, self-hosted" alternative, so I read through the repo, the package scripts, and the release assets to see what you actually get. 1. What the Open Generative AI GitHub repo is The dev.to article that currently ranks for this search is a January 2025 list of three image projects: CompVis Stable Diffusion, DALL-E Mini, and StyleGAN3. Those are model repos. Open Generative AI is a different kind of thing: a front end that puts hundreds of hosted models behind one interface. From the README and package.json: - Stack: Next.js 14 (App Router), React 18, Tailwind CSS v3, npm workspaces. A shared packages/studio component library holds the studios, and packages/studio/src/models.js is the single list of model definitions. - Studios: Image, Video, Audio, Lip Sync, Cinema (camera, lens, focal length, and aperture presets turned into prompt modifiers), Workflow (node-based pipelines), Agent, Design Agent, Marketing, AI Clipping, Body Swap, and a few more. - Distribution: a web build (npm run dev), an Electron desktop build, and a hosted copy on muapi.ai. - License: MIT. The same packages/studio library powers the hosted version on muapi.ai, which explains why the model list moves so fast. Model updates land in one file and ship to both. 2. What's open and what isn't This is the part the repo description ("Self-hosted, MIT licensed") glosses over. | Layer | Where it runs | What you need | | --- | --- | --- | | UI (all studios) | Your machine or server | Node.js 18+, or the desktop installer | | Cloud models (Flux, Kling, Veo, Sora, Seedance, Midjourney...) | Muapi's API | A Muapi access key, billed by Muapi | | sd.cpp local engine | Your machine, desktop app only | CPU works; Metal on Apple Silicon, CUDA/Vulkan/ROCm elsewhere | | Wan2GP local engine | Your GPU box, desktop app only | Your own Wan2GP install on a CUDA or ROCm GPU | | Hosted web version | muapi.ai | A free account; always uses cloud APIs | The API flow is documented in the README. The app submits a job with POST /api/v1/{model-endpoint}, then polls GET /api/v1/predictions/{request_id}/result until the status is completed, authenticating with an x-api-key header. Your key sits in browser localStorage and, per the README, is only sent to Muapi. So "free" means the code costs nothing. Generations on the big commercial models cost whatever Muapi charges for them. That's a perfectly reasonable design, a bring-your-own-key client, but you should budget for it before you tell a stakeholder it's free. 3. Running it from source The README is clear that most people should grab a prebuilt installer. If you want to hack on it, this is the path, and I checked each script name against package.json: # submodules are required for the workflow and agent packages git clone --recurse-submodules https://github.com/Anil-matcha/Open-Generative-AI.git cd Open-Generative-AI # "setup" = git submodule update + npm install + build all workspace packages # plain `npm install` is not enough npm run setup # then ONE of these npm run electron:dev # desktop app (Vite build, then Electron) npm run dev # web version on http://localhost:3000 # production web build npm run build && npm run start If Next.js complains it can't find a pages directory, you're either not in the repo root or you cloned without submodules. Re-run npm run setup if packages/Vibe-Workflow or packages/agents are empty. 4. Local inference: two engines, very different costs Local generation is desktop-only. The hosted and web builds always call cloud APIs. sd.cpp (bundled). Built on stable-diffusion.cpp. You install it with one click under Settings, then Local Models. It handles image models only: Dreamshaper 8, Realistic Vision 5.1, and Anything v5 (SD 1.5, about 2.1 GB each), SDXL Base 1.0 (6.9 GB), and Z-Image Turbo and Base, which also need a 2.4 GB Qwen3-4B text encoder and a 335 MB FLUX VAE. The README warns that Z-Image is known to hang a base 8 GB M-series Mac and recommends 16 GB of RAM. On 8 GB, stick to SD 1.5. Weights default to Electron's app-data folder (~/Library/Application Support/open-generative-ai/local-ai on macOS, ~/.config/open-generative-ai/local-ai on Linux). If you'd rather keep multi-GB weights on another drive, set OPEN_GENERATIVE_AI_LOCAL_AI_DIR before launch. Wan2GP (bring your own). This is where video lives. The app doesn't bundle Python or weights. You install Wan2GP yourself on a CUDA or ROCm machine, and the app either runs wgp.py from a local folder or talks to Wan2GP's MCP server on another box: python wgp.py --mcp --mcp-api-version 1 --mcp-transport streamable-http \ --mcp-host 0.0.0.0 --mcp-port 7866 You paste http://<host>:7866/mcp into the settings panel. The Gradio UI on port 7860 won't work for this. Only bind to 0.0.0.0 on a network you trust. The remote mode is the clever part. Wan2GP has no Apple Silicon path, so a Mac user can keep the desktop app and push inference to a Linux GPU box or a rented instance. Supported models include Wan 2.1 1.3B (480p, about 8 GB VRAM, the lightest option), Wan 2.2 14B T2V/I2V, Hunyuan Video 1.5, LTX-2 Distilled (720p), plus Flux.1 Dev and Qwen Image for stills. 5. Gotchas I'd check before relying on it - The numbers in the README disagree with each other. The tagline says 400+ models, a features bullet says 200+, and the category table says 420+ "verified against models.js." Lip sync is "9 dedicated models" in one section and 15 in the category table. Count models.js yourself if the number matters to you. - The download table is stale. It still links 1.0.9 installers, while the latest GitHub release is v2.0.0 (May 23, 2026), with .dmg builds for arm64 and Intel, a Windows .exe, an .AppImage, and a .deb. Go to the Releases page directly. - The binaries are unsigned. On macOS you'll need xattr -cr "/Applications/Open Generative AI.app" or the "Open Anyway" button; on Windows, SmartScreen's "Run anyway." - Ubuntu 24.04 AppArmor. The AppImage can die silently because of apparmor_restrict_unprivileged_userns. The .deb ships an AppArmor profile, so install that instead of flipping the sysctl. - "No content filters" describes the UI. Your generations still run on Muapi and on the upstream model providers, so read their terms before you build a product on that promise. - The README sells white-label hosting. MuAPI's white-label plan starts at $49/mo. It's optional, but expect upsell links throughout the docs. 6. How it compares to the older GitHub picks | Repo | Stars (Oct 2026) | Last push | What it is | | --- | --- | --- | --- | | Anil-matcha/Open-Generative-AI | 29.7k | Oct 2026 | Multi-model studio UI, BYO key plus local engines | | CompVis/stable-diffusion | 73.5k | Jun 2024 | Original SD research code | | borisdayma/dalle-mini | 14.7k | Nov 2023 | DALL-E Mini model code | | NVlabs/stylegan3 | 6.9k | Sep 2023 | GAN research code, NVIDIA GPU | | deepbeepmeep/Wan2GP | 10.1k | Oct 2026 | Local video/image runtime for modest GPUs | | leejet/stable-diffusion.cpp | 7.5k | Oct 2026 | C++ diffusion inference, powers the sd.cpp engine | The 2025 list is still useful as history. If you want to generate something this week, the three bottom rows are where the activity is, and the top row ties them together behind a UI. FAQ Is Open Generative AI on GitHub really free? The code is MIT licensed and free. Cloud generations go through Muapi with your own key, which is billed. Local generation through sd.cpp or Wan2GP costs only your hardware. Can I run Open Generative AI fully offline? Only in the desktop app, using sd.cpp for images or a Wan2GP install for video. The web build and the hosted version always use cloud APIs. What GPU do I need for local video? Wan2GP needs a CUDA or ROCm GPU. Wan 2.1 1.3B at 480p is listed at about 8 GB VRAM. Larger models such as Wan 2.2 14B are noted as slow on consumer cards. Is Open Generative AI the same as Stable Diffusion? No. Stable Diffusion is a model. Open Generative AI is a studio that can run SD 1.5 and SDXL locally and also calls hundreds of hosted models. Where I landed For prototyping and comparing models side by side, Open Generative AI is the most complete open-source front end I've looked at, and the Wan2GP-over-MCP setup is worth stealing even if you never use the UI. For a production feature where a user clicks "generate" and expects a video back through a webhook, I still prefer a narrower hosted service. For that job I reach for AI Video API, which handles text-to-video and image-to-video with callback webhooks, so my backend stays a queue and a handler. The repo is at Anil-matcha/Open-Generative-AI if you want to read models.js yourself.