- All languages
- Arduino
- BitBake
- C
- C#
- C++
- CMake
- CSS
- Clojure
- CodeQL
- CoffeeScript
- Common Lisp
- Crystal
- Cuda
- Cython
- D2
- Dart
- Elixir
- Emacs Lisp
- Erlang
- F*
- Fortran
- Frege
- GDScript
- Go
- Groovy
- HTML
- Haskell
- Java
- JavaScript
- Jinja
- Jsonnet
- Julia
- Jupyter Notebook
- Kotlin
- Lua
- M4
- MDX
- MLIR
- Makefile
- Markdown
- MiniZinc
- MoonScript
- Nim
- OCaml
- Objective-C
- Objective-C++
- PHP
- Perl
- Pony
- Protocol Buffer
- Python
- Reason
- Rich Text Format
- Ruby
- Rust
- SCSS
- SMT
- Scala
- Scheme
- Shell
- Shen
- Smalltalk
- Svelte
- Swift
- TLA
- TeX
- TypeScript
- Verilog
- WebAssembly
- Wren
- XML
- XSLT
- Zig
Starred repositories
NVIDIA Kubernetes Engine (NVKE). A management plane for GPU Kubernetes on bare metal.
1M param OCR model - runs on CPU, 15x faster than tesseract, close to 1B+ param models in accuracy
DeepInfer: High-performance, large-scale inference framework for DeepSeek models
Turn your regular JavaScript code into a neural net.
Hindsight: Agent Memory That Learns
Jev-compatible API server for open decision models. Ollama for decision models: pull and serve 29 model families (Laya, pplx-decider, Clef, Winnow, coba...) locally or self-hosted behind /v1/system…
Tile-Based Runtime for Ultra-Low-Latency LLM Inference
Contrastive Language Models (CLMs) are a new class of System One model trained with a contrastive learning objective that connects states and actions.
Fast, Flexible and Portable Structured Generation
Jev-like family of decision models built on top of Qwen3.5/3.8 you can train and run on your own
Semantic ifs from open models, on a 3090 at home. Independent; not affiliated with Jev or TypeSafe.
Local typed decisions, contrastive data curation, and model evaluation.
RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing …
Open-source Python library for AI browser agents, plus Browser Use Cloud: stealth browsers and a hosted agent API.
Agent Executor (AX) is Google's open agentic orchestration runtime
NVIDIA's CUDA platform for Rust. Host runtime crates plus Tile (cutile-rs) and SIMT (cuda-oxide) kernel programming models in idiomatic Rust.
SubmitQueue is a high-performance speculative merge queue that keeps your trunk consistently green at scale.
Minimal kernel to make any AI coding agent stateful. Clone, point your agent, go.
Convert PDF to markdown + JSON quickly with high accuracy
ggml speech-to-text inference for 16+ model families
A format specification for describing a visual identity to coding agents. DESIGN.md gives agents a persistent, structured understanding of a design system.
Unlimited OCR Works: Welcome the Era of One-shot Long-horizon Parsing.