Havuz

The best model under 200B at near 10 token/s exact batch throughput on a single laptop CPU, with an automatic 8 GB RAM path. Native C, no GP

Local OCR, metadata automation and RAG document chat for Paperless-ngx using PaddleOCR and Ollama, designed for CPU-only self-hosted servers

Local-first desktop coding agent with BYOK, MCP, and OpenAI-compatible / Ollama hosts

A 100% local AI memory engine for your computer

🔐 Replayable RAG — a local-first, auditable knowledge-reasoning platform. Graph-RAG + append-only audit log + deterministic contradiction ar

Qwen3.8-27B on RTX 5090 Laptop 24GB — TWIN-TURBO NVFP4 (native llama.cpp) vs Coder390 Q3 LynnStyle (.ninfer + NInfer engine in WSL2/Docker):

Writing that's kin to yours

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

Fully local AI companion with Long-Term Memory RAG, branch-isolated worldlines, Live2D/VRM 3D avatars, and real-time TTS for consumer GPUs (

"Review Ducktective" is an automated code review platform based on an LLM: it takes a diff as input and outputs structured findings with sev

Which local LLMs can your computer run? Hardware-aware CLI, TUI and GUI: yes/slow/no verdicts and tok/s from your GPU VRAM and RAM, then ver

Experimental Chrome side-panel browser agent with LangGraph.js, MCP/WebMCP, local models, and semantic browser automation.

MCP server that delegates Claude Code agent work to local models (llama.cpp / vLLM) on your own hardware

MCP server that delegates Claude Code subagents to local models (LiteLLM/llama.cpp/Ollama), DeepSeek, AWS Bedrock, or any OpenAI/Anthropic-c

Deploy Qwen3.6-35B-A3B (Q4_K_XL) + MTP speculative decoding on a single NVIDIA L4 24GB — GCP g2-standard-8 — via the official llama.cpp Dock

A free, private, local AI for any app. One command, OpenAI-compatible, no API key.

Open-source Windows AI IDE built with .NET 10 WPF, featuring a fine-tuned local AI router (1.5B), live Blender & Unity copilots, self-healin

A local-first autonomous agent harness with a web UI, live charts and buttons in its answers, sparks that keep working while you sleep, voic

Local-first macOS AI workspace. Notch-native, 8 providers, your coding agents and their subagents in one place. ⌥Space anywhere.

Local AI harness/workstation: local AI image editor, Ollama image generation routing, SDXL inpainting UI, CivitAI model imports, 8GB VRAM St

Model Dispatch and Memory Allocator: an LLM inference engine for the hardware you own. Custom CPU, ROCm, Vulkan, CUDA, Metal and Level Zero

A memory system for AI agents modeled on human memory: episodic, affective, reconstructive.

Use many on device and cloud models with Apple's Foundation Models framework in iOS and macOS with a full chat experience.

Local desktop coding assistant you can run with any local model. No API key required. It can read your project, edit files, run commands, ve

Use Ollama,OpenRouter, Gemini, cohere and moonshot ai with Codex plugins, MCP tools, tool_search and apply_patch. Use any plugin you want wi

Raise your own AI partner — an open-source, local-first AI agent platform with persistent memory and evolving personality.

Local-first AI coding agent runtime — your models, fail-closed safety, evidence-gated self-improvement

A 6.87 GiB GGUF build of Qwen 3.8 27B — full 27B quality on 8 GB RAM. 96.7% HumanEval, 92.9% MMLU retention, no distillation.

Run models too big for RAM on Apple Silicon — any Hugging Face model, dense or MoE — by streaming weights from disk

VRChat向けのAIエージェント実装です。音声認識、ローカルLLM、VOICEVOX系TTS、VRChat OSC、YOLO/Pose、VLM、長期記憶、感情/欲求モデル、自律行動を組み合わせ、会話だけでなくアバターの移動、視線、表情、エモート、Use/Grab、ピアノ演奏まで

Local agentic AI runtime optimized for CPU-only consumer hardware, tool use and efficient local inference.

AI-assisted rescue boot USB — a Ventoy multi-boot drive on Ubuntu 24.04 with offline LLM inference (Ollama + llama.cpp), pre-installed agent

Local-first, LLM-agnostic memory system — hybrid retrieval (vector + lexical + graph), multi-signal ranking. Runs on 4GB RAM, CPU-only, no c

Catch local LLMs spilling into the CPU. One run shows GPU placement, prefill, decode, memory, power and a verdict.

What I run on my inference server: Qwen3.8-Flash-Next on two RTX 3090s, and how I got here

Private, offline AI coding agent for local Qwen models via Ollama — zero dependencies, real tools, multi-agent. Your code never leaves your

A high-performance Paperclip AI adapter for running local Google Antigravity (agy) agents.

Run a team of AI agents on your own computer. Your files, browser sessions and passwords stay local. Works with Claude Code, Codex, hosted o

MCP vision server for AI agents, Claude, Cursor, OpenCode, VS Code, local LLMs, Ollama, LM Studio, OpenRouter, Gemini, image analysis, and v

Honeycomb Lab — hex map + OpenAI gateway control plane for a home AI fleet

OnDevice Local AI Studio — run LLMs on iPhone offline. SwiftUI workbench with MLX, llama.cpp, whisper.cpp, RAG, and authenticated OpenAI/Ant

Local and cloud GPU interference model orchestrator exposing OpenAI-compatible API from different engines

Native Rust, single-binary MLX inference server for Apple Silicon. OpenAI/Anthropic-compatible LLM serving — text, vision, audio, embeddings

Hosomaki, the intelligence layer for Linux

Local agent infrastructure in one stdlib-only Go binary

Talk to your computer. Icebreaker transforms natural language into safe Linux actions through isolated AI brains, kernel-enforced sandboxes.

Let Claude run code on your real machine — safely — from any Claude chat. Bridges Claude Cowork to Claude Code on your Mac/Linux box. One co

A coding agent that lives in your terminal. Local Ollama first, or hosted models through the Aether AI fleet.

Local-first desktop app for AI agent teams. Runs with Ollama or cloud models. Mac + Windows.