VFX guy doing AI stuff, pushing pixels, Creator @heyglif
Joined June 2024
- Tweets1.6K
- Following119
- Followers128
- Likes6K
srolkk retweeted
Tengo que decir que lo mejor que he descubierto esta semana es poder usar esta voz desde ComfyUI para hacerme toda la documentación de mis nodos. Me lo estoy pasando como un crío creando estos mini tutoriales 😂
srolkk retweeted
👋🏻 Adios al dolor de cabeza al mezclar 3D y VFX con IA en @ComfyUI
Ya tenéis disponible NKD VFX Tools, un pack que recoge muchas de mis herramientas para workflows híbridos sin salir de ComfyUI. Lo mejor de los dos mundos: control de luz, cámara y perspectiva gracias a las técnicas "tradicionales", pero implementadas con un enfoque moderno en un entorno IA.
srolkk retweeted
Watch Glif narrate how it's building a short form version of Odyssey in "Diorama" style.
This is one of many "skills" in Glif, and you can teach it your own formats
1. Build something in chat, using 100+ AI models (from Flux to Seedance)
2. Tell it to lock it in as a skill!
srolkk retweeted
Glif can now inspect its own outputs and self-correct when it finds issues - just tell it to use the verification skill!
Here, it takes a model, an outfit and a simple prompt as inputs, generates, self-corrects and only then puts together the final multi-scene video...
srolkk retweeted
I tested Kimi K3 vs Claude Opus 4.8
Same prompt, an armory bay with lighting, props, and detail. Top is Kimi K3, bottom is Opus 4.8.
It's not even close.
Kimi K3 built a full scene with textures, proper lighting, ammo crates, weapon racks, working detail everywhere. Opus 4.8 gave me a near empty room with a couple of floating tables.
No doubt it beats Opus 4.8. Kimi K3 is Fable 5 level, and it's clearly better than GPT-5.6 Sol at 3D and games.
An open weight model just matched the best closed models on the market. Let that sink in.
srolkk retweeted
shot caller 🌬️⌨️
✏️ @ComfyUI + @AdobeAE 🤖 @Alibaba_Wan 🎨 @krea_ai 2 + @OpenAI GPT Image 2.0 🏎️ @thesystms HUD 🎹 @suno
srolkk retweeted
さらに更新した。ポリゴンリダクションと、エッジのスムース化。
一枚の画像を入力。デプス推定でメッシュのジオラマをつくるツール。
DLするモデルのサイズが1/10に。MITライセンス。
tomosud.github.io/Image_to_M…
srolkk retweeted
Seedream 5.0 Pro is now in ComfyUI via Partner Nodes.
ByteDance's latest image model brings:
→ Character & product consistency across edits
→ Precision region editing
→ Infographics, flowcharts & structured layouts
→ In-image text across 14 languages
One model. Text-to-image, editing, and multi-image inputs.
Click the link below for the workflow 👇
srolkk retweeted
Introducing 📢📢ProxyPose📢📢
✨Track 6-DoF motion of any query pixel via video-to-video translation
📰ArXiv: arxiv.org/abs/2607.06555
💻Code: github.com/ruihangzhang97/pr…
🌐Webpage: ruihangzhang97.github.io/pro…
srolkk retweeted
Architecture building from a single image made with Nano Banana->tripo 3D .glb -> redreamed from multiple angles of the building and stacked, then assembled into video all in one wf within @ComfyUI
