Single-prompt animation test claims full pipeline: modeling, textures, sound
DavidmComfort · x · 2026-11-03
ChrisGPT shared an animation test reportedly made with "Fable 5.5 / Opus 5.6": one prompt, no GPU, no external assets — the model allegedly handled modeling, texture painting, writing its own renderer, 2s animation, effects, and sound synthesis. The model names are not from any known release, so treat as unverified. "Fortiche is still safe… for now," the author jokes.
More from Multimodal
- ChatGPT Pro conjures a surprisingly charming 'Velvet Kin' mascot for an AI agent — RileyRalmuto · 2026-11-03
- Astrocade launches Avatar Studio v1 for creating and exporting game-ready AI avatars — _amirabs · 2026-11-03
- Texturing an entire Blender street with Qwen Image 2.1 via open-source StableGen add-on — sakalond · 2026-11-03
- Solv3R Turns a Looped Transformer Into a Multi-View Geometry Solver With 16× Pipelined Throughput — taiyasaki · 2026-11-03
- OpenAI says 1.5 billion images are generated per week — firstadopter · 2026-11-03
- ComfyUI-LATO2 turns TRELLIS.2/Hunyuan3D meshes into usable low-poly, 990k to 3k faces — sakalond · 2026-11-03