Havuz

The Kimi K3 and DeepSeek-V4-Flash-0731 architectures, reimplemented from scratch in pure Rust (zero dependencies) and verified 1:1 against t

Give Vercel AI SDK agents a real persistent filesystem with Amazon S3 Files.

A self-improving knowledge base about LLM agent infrastructure

End-to-end agentic quantum-ML pipeline: PennyLane GPU quantum-kernel anomaly detection + LangGraph multi-agent system + locally-hosted Qwen3

Why is LLM inference slow — and how do you make it fast? A hands-on, first-principles course: roofline → KV cache → quantization → paralleli

面向 AI 奥林匹克与认证（IOAI / NOAI / LMCC）的在线评测与竞赛平台：支持客观题、代码题、LLM 工程题与产物提交（类 Kaggle）评测。

Open, probabilistic day-ahead load forecasts for the US power grid. Chronos-2 fine-tuned on 7 years of EIA-930 data across all 53 reporting

A beautifully crafted desktop client for running and managing local LLMs via llama.cpp.

100 days, 100 sources: a daily public learning log of modern AI - LLMs, agents, RAG, fine-tuning, MLOps and more. Daily on Threads, X, and B

Real-time airspace surveillance, orbital tracking, and spatial-temporal anomaly detection over live multi-source ADS-B data. Django/Channels

Claw is a local-first AI control plane that runs powerful open models on your machine and connects to top LLM providers. It intelligently ro

A knowledge graph and meta-learning framework for context-aware automated feature engineering

LLM inference engine for AMD RDNA4 — Rust + Vulkan compute shaders, gguf & native FP8.

A structured knowledge base for large language model technologies, covering LLMs, Agents, RAG, model training, evaluation methodologies, and

Transformer inference via unconventional compute: a single algebraic primitive, a tag-system scheduler, and Apple's Neural Engine through co

A state machine protocol for running controlled ML experiments with coding agents.

Reactant-first training for Lux models: declare the experiment, compile once, train without boilerplate.

Fugusashi — Federated Learning of LLM Routing with Human-Interpretable Decisions. Open-source alternative to Sakana AI Fugu.

Runs Hugging Face ONNX models on any Nx backend, with no per-architecture code

This repository is a comprehensive, hands on roadmap to Python Mastery and AI Engineering. It is designed to take a learner from the very fo

Compile your own LLM for the Hailo-10H — an open pipeline from Hugging Face checkpoint to a self-contained HEF, served with hailo-ollama.

Production-ready RAG assistant for Moodle — containerized with Jenkins CI/CD on AWS.

Qwen3.8-27B on one laptop CPU: up to 2.52 token/s, 8 GB tested, no runtime accuracy loss. Native OpenAI-compatible API with function tools;

AI model that animates any SVG into Lottie with 3-4x fewer tokens than OmniLottie (CVPR 2026). Upload SVG → get animated .lottie. 3B model,

How hard can it be to implement a diffusion language model by hand? Easier than I thought, actually.

Illustrated comic-style cheat sheets for ML, LLMs, AI safety, LeetCode and Git, plus a SwiftUI study app (Panels) to read them and quiz your

Discover the citing papers that matter most — ranked by impact, relevance, and influence

Tiny-MoE is a lightweight Mixture-of-Experts language model built entirely from scratch in native PyTorch and trained end-to-end on Kaggle u

Open-source control layer over PDF parsing that picks the right method, verifies output quality, and falls back automatically when results f

A scientific-method runtime for AI agents: hypotheses, experiments, evidence, provenance, reproducibility, and auditable conclusions.

PyTorch-native framework for fault-tolerant distributed training of foundation models on AI clusters

On-device streaming speech-to-text for macOS, built on Qwen3-ASR via MLX

CPU-first, eager-execution tensor/autograd runtime and LLM inference engine written in pure Zig

A free, open GenAI curriculum built from scratch. Start from zero, build real projects, and work your way up to production AI systems.

ThreatRadar is an open-source Threat Intelligence pipeline for SOC and threat intelligence teams that ingests, enriches, scores, and validat

Built for Smart India Hackathon 2026, SIH Problem Statement 26192 by National Disaster Response Force (NDRF), DM Division, Ministry of Home

Self-learning AI agent that gets smarter and cheaper over time. Routes between local and cloud LLMs, learns from every interaction, remember

Asking for What Was Never Requested: horizontal and vertical proactivity in LLM agents. Need-graph metrics (no LLM judge) and Q&D, which tra

A complete, structured end-to-end roadmap to become a versatile Generalist AI Engineer from Python/math foundations through ML/DL, CV, NLP,

Lightweight observability for multi-agent AI pipelines. Self-hosted, no cloud dependency — add two lines of Python and see real-time agent e

Object-storage-native KV cache for LLM inference & RL. Cross-restart, cross-conversation, cross-engine via shared S3 bucket.

🏆 Real-world problem statements for hackathons & impact projects — curated, structured, and ready to build. 65 problems across 5 tracks: Glo

OMCBench is a benchmark suite for evaluating malicious-code detection capabilities. The benchmark consists of a labeled set of 800 Python an

🛡️ Explainable AI Malware Analysis Platform with PE Static Analysis, YARA Engine, SHAP Explainability & Real-Time Dashboard

Local-first AI creative pipeline for persona-based image/video generation. Identity preservation, visual engines, quality scoring, autopilot

🎙️ Adaptive Audio Authentication Engine: Multi-layer voice biometric system with deepfake detection, speaker verification (ECAPA-TDNN), ASR

Tenstorrent backend plugin for vLLM.

Dataset abierto de modelos de HuggingFace: metadatos curados y normalizados, licencias clasificadas, parámetros reales, tumbas de modelos re

🛡️ AI-powered reconnaissance framework unifying 130+ security tools — subdomain enum, vuln scanning, cloud/IoT recon, dark web monitoring &