A blazingly fast, open-source application server with type-safe APIs, built-in WebAssembly runtime, realtime, auth, and admin UI built on Rust, SQLite & Wasmtime.

All-in-one LLM CLI tool featuring Shell Assistant, Chat-REPL, RAG, AI Tools & Agents, with access to OpenAI, Claude, Gemini, Ollama, Groq, and more.

Minimal CLI coding agent by Mistral

A self-distillation based training method for long context reasoning in a single LLM without reinforcement learning

cuTile is a programming model for writing parallel kernels for NVIDIA GPUs

CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning

A general memory system for agents, powered by deep-research

Second iteration of my Rust key–value store — segmented log, in-memory index, checksums, and manual compaction.

Preprint: Asymmetry in Low-Rank Adapters of Foundation Models

dLLM: Simple Diffusion Language Modeling

Fast Diversification for Search & Retrieval

A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.

Official implementation of "Continuous Autoregressive Language Models"

Training and Inference for "Languages are Modalities: Cross-Lingual Alignment via Encoder Injection"

Nearly Inference Free Embeddings: make your RAG queries 500x faster

Ungreedy subword tokenizer and vocabulary trainer for Python, Go & Javascript

Nano vLLM

NanoGPT (124M) in 3 minutes

Automatic differentiation for Triton Kernels

LightMem: Lightweight and Efficient Memory-Augmented Generation

RL gym for vision language models written in JAX

Math-VR Benchmark & CodePlot-CoT: Mathematical Visual Reasoning by Thinking with Code-Driven Images

The best ChatGPT that $100 can buy.

Automatic Video Generation from Scientific Papers

Flask app to interact with FastVLM using CPU only

A Lightweight LLM Post-Training Library

DocLayout-YOLO: Enhancing Document Layout Analysis through Diverse Synthetic Data and Global-to-Local Adaptive Perception