Neuroscience-inspired cognitive substrate: FHRR + Modern Hopfield Networks for contextual completion

A mod for Minecraft Fabric that adds smart PvP bots

Self-hosted AI workspace.

Local UI-grounding specialist for hybrid AI agents. Qwen3-VL-2B LoRA. Screenshot + text target → strict JSON bbox. Drop-in for Claude Code, Codex, browser-use. Cuts GPT-4V cost & latency.

DVOC (Damped Visual Offset Correction) is a training-free, recursive refinement protocol that enables weak vision-language models (VLMs) to achieve pixel-precise UI grounding through iterative visu…

CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies

DFlash: Block Diffusion for Flash Speculative Decoding

~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & En…

A scalable generative AI framework built for researchers and developers working on Large Language Models, Multimodal, and Speech AI (Automatic Speech Recognition and Text-to-Speech)

llama.cpp fork with additional SOTA quants and improved performance

Implementation of Switch Transformers from the paper: "Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity"

Vulkan-based implementation of D3D8, 9, 10 and 11 for Linux / Wine

A lightweight markup language designed with strict, simple rules for reliable text formatting.

A modern, Terminal User Interface (TUI) library for Python with React-inspired hooks, C++ optimized rendering, and built-in virtualization for large datasets.

EBOS: Event-Boosted O(1) Scheduler

Automated scope creep as a service. Stop being productive. Start being tired.

A semantic meta-search engine and synthesis workstation featuring self-correcting vector memory and local AI insights.

Arch-Hyprland dotfiles/distro. A true labor of Passion, Consider rewarding with a star.

KeSSie HUGE Context Semantic recall for Large Language Models

A simple Pomodoro App written in Kotlin. Focus on what matters!

A Proof of Concept for the LEMA (Layer-wise Efficient Memory Abstraction) framework. Enables stable fine-tuning of Llama-2-7B on consumer-grade hardware (16GB VRAM) through layer-wise weight stream…

Know before you train — VRAM estimation for LLM fine-tuning.

LEMA (Layer-wise Efficient Memory Abstraction): A hardware-aware framework for fine-tuning LLMs in VRAM-constrained environments using asynchronous binary pre-fetching and triple-tier memory orches…

Predictive VRAM Virtualization Engine

A list of AI autonomous agents

A complete, isolated internet infrastructure for AI agents. OAuth2 auth, economic system, social network, semantic search, and MCP protocol support.

High-performance daemon for real-time codebase indexing. Generates semantic embeddings locally to provide AI agents with tools for instant search interface.

Early baby steps towards a long-term vision regarding Mamba-2's state interpretability.