ai4anim-webgpu runs batched neural motion matching inference for multiple agents entirely in the browser using WebGPU compute shaders and mixed-precision matmul kernels.
Fun-ASR is Alibaba Tongyi Lab’s end-to-end speech recognition model with 800M parameters, supporting 31 languages and real-time transcription in noisy environments.
IRONSIGHT aggregates 50+ free public sources into a real-time OSINT dashboard with Next.js, combining military tracking, Telegram feeds, market data, and satellite info—all without API keys.
Explore Python-NSE-Option-Chain-Analyzer, a Python tkinter desktop app that fetches and analyzes real-time NSE option chain data with robust polling, deduplication, and error handling.
DAAAM builds real-time 3D dynamic scene graphs using foundation models like SAM and VLMs, targeting large-scale robot mapping with semantic and spatio-temporal memory.
GoAccess is a real-time web log analyzer written in C, offering low-footprint, fast HTML reports with geoip and TLS support. Easy to build or install via packages.
stream-video-js offers a layered TypeScript SDK for building video calls, audio rooms, and livestreams across React, React Native, and vanilla JS. It runs on a global edge network with 99.999% uptime.
MeanVC enables real-time zero-shot voice conversion using mean flows and diffusion transformers for single-step inference, addressing latency bottlenecks in diffusion models.
Explore how mini-tokyo-3d renders Tokyo’s transit network in real-time 3D using live ODPT data and Mapbox GL, with a Node.js build pipeline and multi-language support.
PEAR predicts expressive 3D human mesh parameters for body, hands, and face simultaneously at 100 FPS using a pixel-aligned architecture based on PyTorch and SMPL-X models.
Avatar Forcing implements diffusion forcing for causal, real-time multimodal input processing enabling expressive head avatars with ~500ms latency and 6.8X speedup over baselines.
Vibra Code uses Claude Code in E2B cloud sandboxes with real-time sync and a native iOS chat UI rendering at 60fps off-main-thread. Open source AI app builder with multi-provider support.
Deep-Live-Cam offers real-time face swapping and deepfake video generation using ONNX Runtime with multiple execution providers for optimized performance on GPUs, CPUs, and Apple Silicon.