cyber_phunk retweeted
发现一个硬核的开源 AI 拍电影工具：AI Video Production Editor。
它把剧本、选角、分镜、视频生成、连续性检查、补拍和剪辑整合进了一款桌面软件。
我翻了源码，用官方短片案例跑了一轮独立规则测试：4 个镜头，故意设置 2 处穿帮，全部检出，修正后 4/4 通过。
目前验证的是检查与返工逻辑，原软件的视频自动审查和补拍还没亲测。
比起单纯生成视频，这种完整的 AI 影视制作工作流更值得关注。
GitHub：github.com/LudwigKienle/ai-v…
cyber_phunk retweeted
Just another day as a knight in modern-day France 🇫🇷⚔️
Mission: Find your Mona and protect Louis))
cyber_phunk retweeted
ComfyArt now runs your own ComfyUI workflows, and spreads them across every machine on your LAN.
Your ComfyUI workflow, on any machine you own.
Title 3 nodes, export as API, drop it into ComfyArt. It becomes a model on the canvas.
Add the other ComfyUIs on your network and set Auto: each job goes to the least busy machine that has its models.
No cloud. No credits. Open source.
github.com/jtydhr88/ComfyArt
#OpenSourceHiggsfield #ComfyClub
ComfyArt: an open-source AI design canvas built on ComfyUI.
Run it on your own GPU, or flip one setting and the same workflows run on Comfy Cloud via the Comfy API. No build, no deploy. Open-weights models only.
github.com/jtydhr88/ComfyArt
#OpenSourceHiggsfield @ComfyUI
cyber_phunk retweeted
Stop using video downloader sites full of popups and fake download buttons.
yoinks downloads videos from YouTube, X, Instagram, TikTok, Threads and 1,800+ other sites, right from your terminal. Built by designer Pablo Stanley. 6K stars.
→ paste a link, pick a resolution or audio-only MP3, done
→ no popups, no fake buttons, no sketchy redirects
→ saves straight to your Downloads folder
→ runs on yt-dlp under the hood and sets it all up for you
Try it without installing anything:
npx yoinks
Or install it: npm install -g yoinks
Needs Node 18+. Only download videos you have the right to save.
MIT.
github.com/pablostanley/yoin…
cyber_phunk retweeted
This is too funny 😂
Try out Artlist: artlist.io/artlist-70446/?ar…
Here is your Claude skill: go.genhq.io/artlist-seedance…
Artlist AI runs Seedance 2.5 to turn low-quality footage into cinema, generating brand new camera angles that were never filmed. One clip becomes a full multi-angle sequence, giving creators, filmmakers and advertisers complete directorial control in post.
Our mission is to create 100,000 new jobs for creatives through AI education
Find paid work / AI related jobs→ genhq.io
Learn AI → join.genhq.io/
#ArtlistAI #Seedance #AICameraAngles #AIFilmmaking #CreativeAI
🤝 Paid partnership
cyber_phunk retweeted
nukeにあるflaresimプラグインをAEに移植しました！
githubにオープンソースで公開しています📸✨
github.com/ken-eizo/flaresim…
#aftereffects #flaresim #lensflare
cyber_phunk retweeted
We just open-sourced motion graphics templates.
Each loop is about 30 lines of code your agent can rewrite in your brand colours.
> seamless 12-second backgrounds: depth, lines, dots, shapes
> click a loop to open its code as one HTML file
> a skill your agent uses to design new ones
Comment "MOTION" and I'll send you the library.
cyber_phunk retweeted
Here's a free City Generator that can export to GLB, OBJ, or FBX for your MCP Blender projects.
Even if you're NOT going to use it for 3d scenes, it's just fun to play around with for 5 minutes!
cyber_phunk retweeted
ComfyUI DBS Soundtrack Separation
You can split ANY soundtrack into Music, Sound Effects, and Dialogue with the new DBS models.
- TINY (only 1M–4.7M params)
- run on CPU/GPU
huge for remixing/cleaning up audio.
github.com/hiroki-abe-58/Com…
cyber_phunk retweeted
Turn any Google Street View into a 3D Gaussian splat 🤯
Paste a Google Maps link and turn that spot into an immersive 3D scene you can explore.
Your hometown. Your favorite street. That place you’ve always wanted to visit.
100% free.
cyber_phunk retweeted
Moss Image N$FW 💋💄👠 LoRA
It is tuned for explicit N$FW portraiture, body poses, indoor scenes, and close photographic compositions.
👇
huggingface.co/axiomofmind/M…
cyber_phunk retweeted
"Local Video Editing" is officially dead 🤯
VideoSOS brings 100+ AI models into one browser-based video editing workspace.
- Generate video with Veo 3.1
- Create images with FLUX and Imagen 4
- Use Gemini 2.5 Flash for AI tasks
One interface for your workflow. Processing depends on the model and provider you connect.
100% open source.
cyber_phunk retweeted
ComfyUI MOSS Prompt Enhancer
Turn basic ideas into killer descriptive prompts & exact spatial layouts, running locally with GGUF.
github.com/slmonker/ComfyUI-…
cyber_phunk retweeted
20 Blender 5.2 skills that give AI agents an actual 3D production workflow
The repo covers the full pipeline:
> modeling + sculpting
> retopology + UVs
> PBR materials + baking
> geometry nodes
> rigging + animation
> rendering
> export to Roblox, Unity, Unreal and Godot
The skills are designed for agents like Claude Code, Codex, Cursor, Gemini CLI, Copilot and OpenCode.
There’s also a QA skill that checks for things like z-fighting, floating parts, bad normals and UV issues before an asset is considered finished.
The useful part here is that it gives coding agents a repeatable Blender process instead of forcing them to improvise with bpy every time.
