Stars
Documentation for the 2025 IEEE SSCS Chipathon Track on MOSbius
Mini Mosbius for Tiny Tapeout SKY25A
A language-agnostic package manager with hermetic builds, content-addressable storage, and unified dependency management across ecosystems. Designed for enterprise scale and security
Convert ONNX models into native, backend-agnostic Burn code for inference and fine-tuning.
Open Source framework for voice and multimodal conversational AI
Lightning-Fast, On-Device, Multilingual TTS — running natively via ONNX.
SplitRS uses AST-based analysis to automatically refactor large Rust source files (>1000 lines) into well-organized, compilable modules.
MLX native implementations of state-of-the-art generative image models
Research comparing 6 spec-driven development tools (Spec-Kit, Spec Kitty, BMad, OpenSpec, Kiro, Tessl) with git worktree analysis and decision frameworks
Contract-driven polyglot transpile workbench. Python/C/C++/Rust/Ruchy/Lean ↔ Rust/Ruchy/PTX/WGSL/SPIR-V, with provable contracts at every layer.
VoiRS is a cutting-edge Text-to-Speech (TTS), Voice Recognition, Sound framework that unifies high-performance crates from the cool-japan ecosystem
SPARX: An Open-Source, Automated, Programmatically Generated, Frequency-Scalable Six-Port Receiver in 130-nm CMOS
Open source framework to vibecode and prototype voice agents with Gradium APIs
A curated list of full-duplex spoken dialogue models & benchmarks
MoshiRAG is a compact full-duplex speech language model augmented with asynchronous knowledge retrieval to improve factuality without sacrificing real-time interactivity.
The agent that grows with you
Playwright is a framework for Web Testing and Automation. It allows testing Chromium, Firefox and WebKit with a single API.
LLM inference server with continuous batching & SSD caching for Apple Silicon — managed from the macOS menu bar
Free, open-source web app for learning about ontologies and Microsoft Fabric IQ. Explore a catalogue of pre-built ontologies, design your own visually, export as RDF/XML, and share interactive diag…
Transformer inference via unconventional compute: a single algebraic primitive, a tag-system scheduler, and Apple's Neural Engine through convolution.
Thirteen editorial diagram types for Claude Code. Self-contained HTML + SVG. No shadows, no Mermaid-slop.
Language for high-assurance and high-speed cryptography
Rust library for generating vector embeddings, reranking locally!
Run 70B+ LLMs on Apple Silicon by using SSD as extended memory — intelligent layer streaming and caching for Mac
Gossip-based service discovery (and more) for large distributed systems.