🌊 Images to 3D parallax videos
Run Claude Code 100% on-device with local AI on Apple Silicon. MLX-native Anthropic-API server. 6 fighters incl. Muse-Glimmer 30B (now multimodal — reads images, abliterated), Gemma 4 31B, Qwen 3.5 122B (65 tok/s), DeepSeek V4 Flash (1M ctx). Private, offline, airgap-ready. Built for NDA / legal / healthcare workflows.
Code to accompany our paper Chen and Zimmermann (2020), "Open source cross-sectional asset pricing"
[ECCV 2026] Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagery
An immersion toolkit for learning Languages through games and other visual media.
Piano LED Visualizer: Connect an LED strip to your Raspberry Pi and create an immersive visual experience for your piano playing
Up to 4× faster LLM decoding on Apple Silicon, lossless. Native MLX port of DeepSeek's DSpark & z-lab's DFlash speculative decoding — Gemma-4, Qwen3.8, Muse-Glimmer, Nemotron, LFM2.5, Ornith-1.0, ternary Bonsai-27B.
Laplace approximations for Deep Learning.
Open recipes, engine patches, and benchmark harnesses for LLM inference on Intel Arc Pro B60/B70 (Battlemage, Xe2). MoE 35B at 160 t/s decode / 7.5K t/s prefill single-stream, 27B at 50~ t/s decode / 1.7K t/s prefill single stream. vLLM XPU MTP unlocked. Muse Glimmer recipe added!!
面向互动游戏与具身智能的 沉浸式风格主动异步震动反馈Skill 通过异步通信桥连接ESP32-S3蓝牙驱动低功率震动设备 Agent可依据任务进度或错误选择和剧情节点主动发送强度或节奏编排 还会将任务或游戏事件与Agent情绪转化为可感知的震动
One learned latent for geometry + physics + manufacturability: give it a box and a load, get a solver-certified bracket. SE(3)-equivariant field model, immersed FEA, live 3D designer.