WanGP 12.71 ships Viggle-Animate: swap any character into videos in 3 local steps
cocktailpeanut · x · 2026-09-07
Developer cocktailpeanut tested WanGP 12.71 and says quality "completely exceeded expectations." The update integrates Viggle-Animate: no pose/mask/face/depth workflows needed — one reference frame plus 3 steps swaps any character into a video on a local PC. Other highlights: DLSS5 neural rendering with up-to-6x temporal upsampling; expanded Minimax H3 with two-phase generation, inpainting/outpainting, audio refinement and 8-step high-quality PDD; Sense Nova U1.5, an image generator/editor strong at rendering text and infographics; and Deepy, an assistant now powered by a fast Qwen 3.8 27B that can analyze videos, sort photo collections, and even plug into Codex or Claude.
More from Multimodal
- FLUX.3 Image Klein tested across 15 complex prompts: strong on text, weak on distant faces — SkirtSwimming8950 · 2026-10-23
- Two Midjourney sref codes for grainy pastel fashion illustration style — sergeantsref · 2026-10-23
- Three free ways to try MOSS-Image: API playground, HF Space, ComfyUI — songguoxiansen · 2026-10-23
- Open-source MOSS-Image tops 1M ComfyUI downloads, 5B model rivals closed rivals — songguoxiansen · 2026-10-23
- Testing light effects in MiniMax video generation — the guinea pig was fine — seppe0815 · 2026-10-23
- GAFU Uses Generation-Side Representations to Boost Fine-Grained Visual Understanding in Unified Multimodal Models — Tengju Ru · 2026-10-23