Reddit dev asks: which ComfyUI video model (Wan 2.2?) to rent cloud GPU for anime series
Blaze-Morpheus · reddit · 2026-10-13
A software developer with dissociative identity disorder wants to turn a story he has built over four years into an anime series and asks for model advice: with only an RTX 3060 (12GB VRAM), he plans to rent cloud GPUs on Vast.ai and work in ComfyUI. He's weighing Wan 2.2 against alternatives, ideally wanting video+dialogue generation or at least lipsync, with manual editing or eventual automation — being a dev tired of the 9-5 himself.
More from Multimodal
- Ideogram 4.5 lands on invideo, pitched as the most precise image edit model — aziz4ai · 2026-10-13
- Runway co-CEO essay: from pixel simulator to a universal simulator of reality — notiansans · 2026-10-13
- One artist, 30 characters, five seasons: AI-assisted animated IP 'The Chronicles of Bone' — Kavanthekid · 2026-10-13
- RVM: reward-based velocity matching fine-tunes diffusion models with RL without likelihood ratios — ArashVahdat · 2026-10-13
- Muse-Wonder vs Fable-5.1: same-prompt 3D scene generation comparison — adonis_singh · 2026-10-13
- MiniMax H3 video demo: rose petals drift together to spell LOVE — azed_ai · 2026-10-13