Releases

Idiomatic Rust binding to the RNP OpenPGP C FFI (librnp)

Terminal renderer for librio-vt render state, powered by sugarloaf

Backend infrastructure for Rio terminal

Sugarloaf is Rio rendering engine, designed to be multiplatform. It is based on WebGPU, Rust library for Desktops and WebAssembly for Web (JavaScript). This project is created and maintained for Rio terminal purposes but feel free to use it.

Core library for tree-sitter language pack - provides compiled parsers for 306 languages

Internal headless-browser fallback used by the crawlberg crawler. Not intended for direct use.

Universal LLM API client with Rust-powered polyglot bindings.

Candle-based VLM OCR engines for Xberg - pure-Rust transformer OCR (TrOCR, PaddleOCR-VL, GLM-OCR)

Backend infrastructure for Rio terminal

Terminal renderer for librio-vt render state, powered by sugarloaf

Sugarloaf is Rio rendering engine, designed to be multiplatform. It is based on WebGPU, Rust library for Desktops and WebAssembly for Web (JavaScript). This project is created and maintained for Rio terminal purposes but feel free to use it.

Rust bindings for Tesseract OCR with cross-compilation, C++17, and caching improvements

MLX-native ONNX Runtime execution provider (plugin EP) for Apple Silicon — binds mlx-c directly, no mlx-rs.

Sugarloaf is Rio rendering engine, designed to be multiplatform. It is based on WebGPU, Rust library for Desktops and WebAssembly for Web (JavaScript). This project is created and maintained for Rio terminal purposes but feel free to use it.

Async Rust library for driving Android devices in tests via Maestro's gRPC server

YAML Editing and Refactoring with Better Accuracy

P2P mesh networking for your devices, built on Tailscale

Auto-downloads the truffle mesh networking sidecar binary at build time

OpenAI-compatible HTTP server for onnx-genai

Rust inference runtime for generative AI models on ONNX Runtime

Text generation engine combining ONNX Runtime, scheduling, and KV caching

Safe Rust APIs for generative model inference with ONNX Runtime

C ABI for the ORT 2.0 runtime: a libonnxruntime.so drop-in surface (skeleton)

Session and inference API for the ORT 2.0 runtime: intent-based SessionBuilder and sequential executor (skeleton)

A utility crate for mapping and converting Nahpu project data into Darwin Core JSON format.

Pure Rust ML inference engine built from scratch - model serving for GGUF and safetensors

CUDA execution provider for the ORT 2.0 runtime (Phase 2a: cudarc + cuBLASLt MatMul; custom fused kernels deferred)

Model, Data and Recipe Registry with full lineage tracking

Eager (single-op) execution engine for the ORT 2.0 runtime: dispatch individual ONNX ops to EP kernels without building a graph (docs/EAGER.md)

Pure Rust system call tracer with source-aware correlation for Rust binaries