AI Video2024
Generates video from text or images via a DiT-based latent diffusion model: text-to-video, image-to-video, frame extension, and multi-keyframe conditioning in one model. A distilled 2B variant runs near real-time on one H100; 13B for higher quality.
AI Video2025
Turns a raw idea, novel, or screenplay into a complete multi-shot video through a multi-agent pipeline that scripts, storyboards, and renders shots while a vision model checks character and scene consistency across the whole story.
AI Client2025
Runs open-source LLMs and multimodal models entirely on mobile devices for offline, private inference. Offers Agent Skills, Thinking Mode, Ask Image, audio scribe, model management and benchmarks, with Gemma 4 and Hugging Face integration.
AI Agent2025
Demonstrates orchestration of specialist customer-service agents built with the OpenAI Agents SDK, pairing a Python backend for agent logic with a Next.js UI (ChatKit) to visualize routing, guardrails, and demo flows. Useful for prototyping multi-agent customer-service workflows; uses mock flight data and requires an OpenAI API key.
AI Image2025
A 20B-parameter MMDiT diffusion model that generates and edits images with accurate embedded text, including dense Chinese and English typography. Handles complex multi-line layouts and identity-preserving edits while keeping text legible.
AI Client2025
Generate parametric 3D CAD models from natural language and images in the browser, with real-time preview and exports to STL/SCAD. Runs client-side via OpenSCAD WebAssembly, extracts adjustable parameters, and integrates Anthropic for shape generation—suited for rapid prototyping and 3D-printing workflows.
AI Agent2025
Eight example apps for building with the Claude Agent SDK: an IMAP email assistant, a multi-agent research system, an Excel agent, a React/WebSocket chat UI, a .docx resume generator, and hello-world session demos. Local-only, not production.
AI Model2025
Performs open-vocabulary promptable segmentation and tracking on images and videos, returning instance masks and unique IDs from short noun-phrase or exemplar prompts. Trained on a 4M-concept, 52M-mask dataset; uses a presence head and a decoupled detector–tracker design, with near real-time inference for multi-object video.
AI Video2026
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.
AI Model2026
Generates anime-style and other non-photorealistic illustrations from text prompts. A 2B-parameter diffusion base preview trained on millions of anime images (and ~800k non-anime art) and released under a non-commercial license; best used in ComfyUI around ~1MP resolution.
Embodied AI2026
Enables research-grade character animation with neural networks in a single NumPy/PyTorch environment — train models, run inference, and visualize results without leaving Python. Includes ECS-style architecture, mocap import (GLB/FBX/BVH), built-in renderer, and headless/standalone modes for rapid prototyping.
AI Infra2026
Measures multiline text layout and block height without triggering browser reflow: it measures text segments once via Canvas+Intl.Segmenter and caches widths, then computes line breaks with pure arithmetic. Useful for streaming AI text, virtualization, and custom per-line rendering.