Production-ready implementation of InvisPose - a revolutionary WiFi-based dense human pose estimation system that enables real-time full-body tracking through walls using commodity mesh routers

A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.

A real-time streaming conversational video system that transforms text interactions into continuous, high-fidelity video responses using autoregressive diffusion.

tiny vision language model

Open Vision Agents by Stream. Build Vision Agents quickly with any model or video provider. Uses Stream's edge network for ultra-low latency.

Speech-to-speech AI assistant with natural conversation flow, mid-speech interruption, vision capabilities and AI-initiated follow-ups. Features low-latency audio streaming, dynamic visual feedback…

Implementation of "Live Avatar: Streaming Real-time Audio-Driven Avatar Generation with Infinite Length"

very good whiteboard infinite canvas SDK

👨‍🎨 The ergonomic way to storyboard. Turns sketches and annotations into videos by drawing on a canvas.

StreamDiffusion: A Pipeline-Level Solution for Real-Time Interactive Generation

MotionStream: Real-Time Video Generation with Interactive Motion Controls

🎒 Token-Oriented Object Notation (TOON) – Compact, human-readable, schema-aware JSON for LLM prompts. Spec, benchmarks, TypeScript SDK.

"ViMax: Agentic Video Generation (Director, Screenwriter, Producer, and Video Generator All-in-One)"

A tiny CPU simulator written in Python

[ICLR 2024] Efficient Streaming Language Models with Attention Sinks

StreamingVLM: Real-Time Understanding for Infinite Video Streams

The best ChatGPT that $100 can buy.

A comprehensive toolkit for reliably locking, packing and deploying environments for ComfyUI workflows.

A fast AI Video Generator for the GPU Poor. Supports Wan 2.1/2.2, Qwen Image, Hunyuan Video, LTX Video and Flux.

Reached #13 on Stanford's Terminal Bench leaderboard. Orchestrator, explorer & coder agents working together with intelligent context sharing.

for rileys podcast

AG-UI: the Agent-User Interaction Protocol. Bring Agents into Frontend Applications.

Kortix – build, manage and train AI Agents.

[NeurIPS 2025] Image editing is worth a single LoRA! 0.1% training data for fantastic image editing! Surpasses GPT-4o in ID persistence~ MoE ckpt released! Only 4GB VRAM is enough to run!

[CVPR 2025] Official PyTorch implementation of "EdgeTAM: On-Device Track Anything Model"

Whisper-Flow is a framework designed to enable real-time transcription of audio content using OpenAI’s Whisper model. Rather than processing entire files after upload (“batch mode”), Whisper-Flow a…

Lets make video diffusion practical!

💡 VideoMind: A Chain-of-LoRA Agent for Long Video Reasoning

A CPU Realtime VLM in 500M. Surpassed Moondream2 and SmolVLM. Training from scratch with ease.