Starred repositories
Sparse autoencoders for matrix-valued recurrent states in GatedDeltaNet (Qwen3.5, DeltaNet, Mamba-2). Bilinear encoder + flat decoder wins under state replacement.
An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.
100M tokens. Infinite compute. Lowest val loss wins.
Official implementation of PRISM: Deriving the Transformer as a Signal-Denoising Operator
Audio playback and capture library written in C, in a single source file.
A Unified Framework for High-Performance and Extensible LLM Steering
🐉 Loong: Synthesize Long CoTs at Scale through Verifiers.
The nnsight package enables interpreting and manipulating the internals of deep learned models.
Context7 Platform -- Up-to-date code documentation for LLMs and AI code editors
Diffusion Language Models For Code Infilling Beyond Fixed-size Canvas
FULL Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI…
Fish Detection (Segmentation) & Classification models and training scripts
An awesome & curated list of best LLMOps tools for developers
Witness the aha moment of VLM with less than $3.
[ ICCV 2025 ] FaceXFormer: A Unified Transformer for Facial Analysis
A lightweight Database GUI in your browser. It supports connecting to Postgres, MySQL, and SQLite.
Release repo for our SLAM Handbook
High-availability network proxy / VPN server, powered by WireGuard
Make a cascading timeline from markdown-like text. Supports simple American/European date styles, ISO8601, images, links, locations, and more.
Open-source evaluation toolkit of large multi-modality models (LMMs), support 220+ LMMs, 80+ benchmarks
【EMNLP 2024🔥】Video-LLaVA: Learning United Visual Representation by Alignment Before Projection