Tag
Explore by tags
Runs text-to-video, image-to-video, text-to-image, and image editing inference with acceleration, offloading, quantization, and distributed execution for large visual generation models.
Generates real-time, infinite-length portrait video from one reference image on a 12GB GPU. Combines implicit facial signals and 3D keypoints with step-distilled diffusion and autoregressive micro-chunk streaming for low-latency live use.
Generates summaries from URLs, YouTube videos, podcasts, PDFs, and local audio or video files. Backend-agnostic by design: the same pipeline drives local coding CLIs (Claude, Codex, Gemini) or hosted API providers (OpenAI, Google, xAI).
Large-scale, real-world dual-arm video corpus for embodied robotics and reinforcement-learning research — over 1TB of multimodal recordings on Hugging Face, intended for training and evaluating agents in real manipulation scenarios; CC BY‑SA 4.0.
Provides a DiT-based audio–video foundation model plus an official Python inference and LoRA trainer. Ships multiple production-ready pipelines (text/image/audio→video), checkpoints, and performance optimizations (FP8, distilled pipelines) for high-fidelity synchronized audio–video generation.
Browser-based, client-side video editor for multi-track editing, GPU-accelerated preview and local exports without uploading files; leverages WebCodecs/WebGPU and includes an AI upscaling option.
Reduces object-driven shortcut learning in zero-shot compositional action recognition by enforcing temporal verb cues and regularizing against frequent object-verb co-occurrence priors. Proposes RCORE with Co-occurrence Prior Regularization (treats frequent co-occurrences as hard negatives) and Temporal Order Regularization. Evaluated on Sth-com and EK100-com with improved compositional generalization.
Unmixes green‑screen pixels with a neural model to recover straight (unmultiplied) foreground color and a clean linear alpha for every pixel, preserving hair, motion blur and translucency. Produces VFX‑standard EXR outputs, supports optional AlphaHint generators (GVM/VideoMaMa) and Docker/consumer‑GPU optimizations.
ComfyUI workflows that run LTX‑2.3 split models to produce text→video, image→video and audio→video pipelines. Uses extracted/split safetensor or GGUF files so models load more modularly; requires up‑to‑date ComfyUI, KJNodes and ComfyUI‑GGUF.
Author HTML-based video compositions and render deterministic, frame-accurate MP4s with agent-friendly tooling — preview in the browser, drive generation via AI agent skills, and use adapter runtimes (GSAP, Lottie, Three.js).
Provides an annotated multimodal human-motion dataset for language-to-action and robotics research, with BVH and MuJoCo files plus recordings targeted at Unitree-G1 and NVIDIA-SOMA platforms. Covers locomotion, gestures, dance and object interaction with English annotations and 100K–1M samples.
Generate text, images, video, audio and action/robot trajectories from combined text, image, video, audio and action inputs. A Mixture-of-Transformers omnimodal foundation model (Cosmos3‑Nano, 16B params) focused on Physical AI (robotics, AV, simulation) and optimized for NVIDIA GPU runtimes.