DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling

[NeurIPS 2025] TalkCuts: A Large-Scale Dataset for Multi-Shot Human Speech Video Generation

AI-powered job search system built on Claude Code. 14 skill modes, Go dashboard, PDF generation, batch processing.

Claude Code skill for generating talking head videos using VEED Fabric 1.0 via fal.ai

Train the smallest LM you can that fits in 16MB. Best model wins!

Helios: Real Real-Time Long Video Generation Model

The official SpeakerVid-5M data curation code.

SoulX-FlashTalk is the first 14B model to achieve sub-second start-up latency (0.87s) while maintaining a real-time throughput of 32 FPS on an 8xH800 node.

This tool retrieves tokens for all devices connected to Xiaomi cloud and encryption keys for BLE devices.

GDI, OpenGL and Direct3D 9 re-implementation of the DirectDraw API for classic 2D games for better compatibility with Windows ME, 2000, XP, Vista, 7, 8, 10, 11, Wine (Linux/macOS/Android) and Virtu…

The Unofficial TikTok API Wrapper In Python

The real state 10k dataset from https://google.github.io/realestate10k

More relighting!

iikira/BaiduPCS-Go原版基础上集成了分享链接/秒传链接转存功能

Wan2.2-Lightning: Speed up wan2.2 model with distillation

High-throughput tensor loading for PyTorch

HuMo: Human-Centric Video Generation via Collaborative Multi-Modal Conditioning

Speed up model training by fixing data loading.

Complete solutions to the Programming Massively Parallel Processors Edition 4

https://wavespeed.ai/ Context parallel attention that accelerates DiT model inference with dynamic caching

​​Unlimited-length talking video generation​​ that supports image-to-video and video-to-video generation

We present StableAvatar, the first end-to-end video diffusion transformer, which synthesizes infinite-length high-quality audio-driven avatar videos without any post-processing, conditioned on a re…

TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. Tensor…

macOS command-line utility for retrieving currently playing media

[NeurIPS 2025] Let Them Talk: Audio-Driven Multi-Person Conversational Video Generation

Automatically ⬆️ upload TikTok videos

Unofficial extension implementation of Self-Forcing to support I2V && 14B training.