Viggle Animate is now open source! 🔥
driving video + frame of what to animate = animated video. in 4 steps, H3 base
▶️ on Spaces huggingface.co/spaces/Viggle/…
LLaDA Image Turbo 6B is here! A 4-step distilled version of LLADA Image, an unified diffusion model: both the text-encoder & the transformer are diffusion models 🧨
It is a competitive model for its size, matching models like Z-Image on benchmarks
▶️ huggingface.co/spaces/hugging…
StreamTalk is here: audio in, body gesture out, and you get the pose skeleton (SMPL-X), so the 3D rig is up to you 🗣
▶️ on Spaces hf.co/spaces/hugging…
H3-World is out on Hugging Face 🎮
a LoRA that turns MiniMax-H3 into a "world model" or a video model conditioned on playable button presses: hold WASD to walk, JKL to swing the camera
▶️ on Spaces hf.co/spaces/hugging…
Krea2-SDA LoRA fixes the biggest issue with Krea2 Turbo: the lack of variety on the same prompt, even with different seeds
by applying semantic directional alignment in 2 of the 8 generation steps, we get way more diversity of outcomes
▶️huggingface.co/spaces/hugging…
