██████╗ ██████╗ ███████╗ ███╗   ██╗
                 ██╔═══██╗██╔══██╗██╔════╝ ████╗  ██║
                 ██║   ██║██████╔╝█████╗   ██╔██╗ ██║
                 ██║   ██║██╔═══╝ ██╔══╝   ██║╚██╗██║
                 ╚██████╔╝██║     ███████╗ ██║ ╚████║
                  ╚═════╝ ╚═╝     ╚══════╝╚═╝  ╚═══╝
███╗   ███╗██╗   ██╗████████╗██╗  ██╗ ██████╗ ███████╗       ██████╗
████╗ ████║╚██╗ ██╔╝╚══██╔══╝██║  ██║██╔═══██╗██╔════╝       ╚════██╗
██╔████╔██║ ╚████╔╝    ██║   ███████║██║   ██║███████╗  █████╗ █████╔╝
██║╚██╔╝██║  ╚██╔╝     ██║   ██╔══██║██║   ██║╚════██║  ╚════╝██╔═══╝
██║ ╚═╝ ██║   ██║      ██║   ██║  ██║╚██████╔╝███████║       ███████╗
╚═╝     ╚═╝   ╚═╝      ╚═╝   ╚═╝  ╚═╝ ╚═════╝ ╚══════╝       ╚══════╝

> *"From the void of computation, a new oracle speaks — no gods, no cloud, only the terminal."*


[](https://www.npmjs.com/package/open-mythos-2) [](https://github.com/creatorofsomethingthatisgood/Open-Mythos-2/stargazers) [](https://github.com/creatorofsomethingthatisgood/Open-Mythos-2/network/members) [](LICENSE) []() []() []() []()
**Fully local, offline AI in your terminal. Zero API keys. Zero cloud. Zero limits.** [Get started in 30 seconds](#-getting-started) · [Features](#-features) · [Commands](#-command-reference) · [RML](#-rml--reinforcement-machine-learning) · [Contributing](#-contributing)

---

Why OpenMythos-2?

	Cloud AI (ChatGPT, Claude, etc.)	OpenMythos-2
Privacy	Your data leaves your machine	Everything stays local
Cost	Monthly subscription or per-token fees	Free forever
Offline	Requires internet	Works anywhere
Setup	API keys, accounts, billing	Just run it
Customization	Locked persona & behavior	RML adapts to you
Security scanning	Separate tool needed	Built-in SAST + AI audit

---

⚡ What is OpenMythos-2?

OpenMythos-2 is a fully local, offline AI that lives and breathes inside your terminal. Inspired by the ancient myths of oracles, gods, and forgotten wisdom — it brings the power of intelligent conversation and reasoning directly to your command line, with zero API keys, zero cloud dependency, and zero limits.

Like Prometheus stealing fire from the gods, OpenMythos-2 brings the fire of AI to every machine.

---

🏛️ Features

Feature	Description
No API Required	Fully local — nothing ever leaves your machine
Terminal Native	Built from the ground up for the command line
Intelligent Reasoning	Context-aware, multi-turn conversations
Offline First	Works anywhere — planes, bunkers, the underworld
Mythos Persona	Answers in the voice of an ancient, wise oracle
Lightweight	Minimal dependencies, blazing fast startup
Open Source	Fully transparent and community-driven
Private by Design	Your conversations are yours alone
RML (Reinforcement ML)	Learns from your feedback and adapts — edits system prompt and generation params to match your preferences	when command is /rml on

The Myth Behind the Machine

The name draws from mythos (μῦθος) — the ancient Greek word for story, legend, and the spoken word of truth. OpenMythos-2 embodies that spirit: a storyteller, reasoner, and companion that runs entirely on your machine, with no external gods (servers) to pray to.

---

Getting Started

Prerequisites

• Python 3.10+
• pip
• A terminal (bash, zsh, PowerShell, cmd)

Installation

Pick your platform and package manager:

---

Linux — Git Clone (recommended)

**Ubuntu / Debian:** ```bash # Clone the sacred repository git clone https://github.com/creatorofsomethingthatisgood/Open-Mythos-2.git cd Open-Mythos-2 # One-time setup (installs deps, builds llama-cpp-python with Vulkan/CPU fallback) sudo bash setup.sh # Download the model (~4.5 GB, first time only) mythos model download # Start chatting mythos ``` > **Note:** `./setup.sh` is optimized for Fedora (`dnf`). On Ubuntu/Debian, it will warn but still work — system deps may need manual install: > ```bash > sudo apt install -y python3 python3-pip python3-dev gcc g++ make cmake git libopenblas-dev libvulkan-dev mesa-vulkan-drivers > ``` **Arch Linux:** ```bash git clone https://github.com/creatorofsomethingthatisgood/Open-Mythos-2.git cd Open-Mythos-2 # Install system deps first sudo pacman -S --needed python python-pip gcc make cmake vulkan-headers vulkan-icd-loader openblas # Run setup sudo bash setup.sh mythos model download mythos ``` **Fedora:** ```bash git clone https://github.com/creatorofsomethingthatisgood/Open-Mythos-2.git cd Open-Mythos-2 # One-command setup (Fedora-optimized, includes Vulkan GPU support) sudo bash setup.sh mythos model download mythos ``` **Offline / Air-gapped machines:** ```bash # On a connected machine: ./scripts/mythos-export-data.sh # creates offline-bundle/ (~4.5 GB) # Transfer the bundle to the target machine, then: ./scripts/mythos-import-data.sh ./offline-bundle ```

macOS — Git Clone (recommended)

```bash # Clone the sacred repository git clone https://github.com/creatorofsomethingthatisgood/Open-Mythos-2.git cd Open-Mythos-2 # One-time setup (Metal GPU on Apple Silicon, CPU on Intel Macs) ./setup-macos.sh # Download the model (~4.5 GB, first time only) ./mythos model download # Start chatting ./mythos ``` > **Note:** On Apple Silicon (M1/M2/M3/M4), the setup builds llama-cpp-python with **Metal GPU acceleration** for fast inference. Intel Macs fall back to CPU mode. [Homebrew](https://brew.sh/) is recommended for build dependencies (`brew install cmake`). **via Homebrew (tap):** ```bash brew tap creatorofsomethingthatisgood/tap brew install open-mythos-2 mythos model download mythos ```

Windows — Git Clone (recommended)

```powershell # Clone the sacred repository git clone https://github.com/creatorofsomethingthatisgood/Open-Mythos-2.git cd Open-Mythos-2 # One-time setup (PowerShell — detects CUDA/Vulkan/CPU automatically) .\setup-windows.ps1 # Download the model (~4.5 GB, first time only) .\mythos.bat model download # Start chatting .\mythos.bat ``` **Prerequisites:** - **Python 3.10+** — [python.org](https://www.python.org/downloads/) (check "Add Python to PATH" during install) - **Microsoft C++ Build Tools** — [visualstudio.microsoft.com](https://visualstudio.microsoft.com/visual-cpp-build-tools/) (select "Desktop development with C++" workload) - **CMake** — `winget install Kitware.CMake` or [cmake.org](https://cmake.org/download/) - **GPU acceleration (optional):** - **NVIDIA:** Install [CUDA Toolkit](https://developer.nvidia.com/cuda-downloads) for CUDA support - **AMD/Intel:** Install [Vulkan SDK](https://vulkan.lunarg.com/sdk/home) for Vulkan support > **Note:** If you skip the build tools, the setup will attempt a prebuilt wheel (CPU-only). For GPU acceleration, the C++ build tools are required.

---

npm (cross-platform: Windows, macOS, Linux)

```bash # Install globally sudo npm install -g open-mythos-2 # Download the model (~4.5 GB, first time only) mythos model download # Start chatting mythos ``` > **Note:** The npm package wraps the same setup and Python backend under the hood. **Node.js 18+** and **Python 3.10+** are required. The first `mythos` run will automatically set up the virtual environment and dependencies if they aren't already present. On Windows, make sure Python is in your PATH.

pnpm (cross-platform: Windows, macOS, Linux)

```bash # Install globally pnpm install -g open-mythos-2 # Download the model (~4.5 GB, first time only) mythos model download # Start chatting mythos ``` > **Note:** [pnpm](https://pnpm.io/) is a fast, disk-efficient package manager. The package wraps the same setup and Python backend under the hood. Node.js 18+ and Python 3.10+ are required. The first run will automatically set up the virtual environment and dependencies if they aren't already present.

pipx (Python users — cross-platform)

```bash # Install with pipx (isolated environment, no venv management needed) pipx install mythos-sentinel # Or install from source git clone https://github.com/creatorofsomethingthatisgood/Open-Mythos-2.git cd Open-Mythos-2 pipx install . # Download the model (~4.5 GB, first time only) mythos model download # Start chatting mythos ``` > **Note:** [pipx](https://pipx.pypa.io/) installs Python CLI tools into isolated virtualenvs. You'll still need the C++ build tools (see platform-specific sections above) for `llama-cpp-python` to compile with GPU support. If building from source, `pipx install .` uses `pyproject.toml`.

Docker (cross-platform — no local deps needed)

```bash # Pull and run (CPU-only) docker run -it --rm \ -v mythos-data:/root/.config/mythos \ ghcr.io/creatorofsomethingthatisgood/open-mythos-2:latest # With NVIDIA GPU acceleration docker run -it --rm --gpus all \ -v mythos-data:/root/.config/mythos \ ghcr.io/creatorofsomethingthatisgood/open-mythos-2:latest-cuda # Download model first (one-time) docker run -it --rm \ -v mythos-data:/root/.config/mythos \ ghcr.io/creatorofsomethingthatisgood/open-mythos-2:latest \ mythos model download # Web UI docker run -it --rm -p 7860:7860 \ -v mythos-data:/root/.config/mythos \ ghcr.io/creatorofsomethingthatisgood/open-mythos-2:latest \ mythos web --port 7860 ``` > **Note:** Docker avoids the need for local C++ build tools. The CUDA image requires the [NVIDIA Container Toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html). Model data persists in the `mythos-data` Docker volume.

⚡ Quick comparison

| Method | Platforms | GPU | Best for | |--------|-----------|-----|----------| | **Git Clone** | Linux, macOS, Windows | Vulkan / Metal / CUDA | Full control, offline setups | | **npm / pnpm** | Linux, macOS, Windows | Auto-detect | JS devs, quick install | | **pipx** | Linux, macOS, Windows | Manual CMAKE_ARGS | Python users | | **Docker** | Linux, macOS, Windows | CUDA (NVIDIA toolkit) | Reproducible, no local deps | | **Homebrew** | macOS only | Metal (auto) | Mac users who prefer brew |

---

RML — Reinforcement Machine Learning

RML is a feedback-driven self-improvement loop that learns your preferences over time and adapts Mythos to match. It does two things automatically:

1. Edits the System Prompt

When a category accumulates enough negative feedback, RML injects a concrete behavioral hint directly into the system prompt. These learned hints tell Mythos how to adjust its style — for example:

• Accuracy hint: “Prioritize accuracy over creativity. Double-check facts. If unsure, say so.”
• Conciseness hint: “Be concise. Get to the answer quickly; elaborate only when asked.”
• Clarity hint: “Use headers, bullet points, short paragraphs. Avoid jargon without explanation.”
• Code quality hint: “Write production-quality code: type hints, docstrings, error handling.”
• Security hint: “Apply security best practices: input validation, no hardcoded secrets.”
• Completeness hint: “Be thorough. Address all parts of the question. Don’t skip edge cases.”

Hints are removed automatically when the category’s score recovers — so the system prompt is always a live reflection of what you actually prefer.

2. Adjusts Generation Parameters

RML also tweaks temperature, top_p, and repeat_penalty behind the scenes:

• High accept rate → temperature nudges up (more creative/confident)
• High rejections/edits → temperature nudges down (more conservative/precise)
• Many interrupts → repeat_penalty increases slightly

All adjustments are bounded by max_param_offset (default 0.3) so nothing swings wildly.

Feedback Signals RML Collects

Signal	Strength	Trigger
Explicit good	+2.0	/rml good
Explicit bad	−2.0	/rml bad
Implicit positive	+1.0	You say “thanks”, “perfect”, “works”, etc.
Implicit negative	−1.0	You say “wrong”, “try again”, “mistake”, etc.
Edit penalty	−1.0	You rewrite Mythos’ output
Interrupt penalty	−0.5	You Ctrl+C during generation

Commands

/rml on          Enable RML
/rml off         Disable RML
/rml good        Mark last response as good (explicit +2)
/rml bad         Mark last response as bad (explicit -2)
/rml stats       Show what RML has learned (scores, hints, param adjustments)
/rml reset       Wipe all learned preferences and start fresh

Config (config.yaml)

rml:
  enabled: false          # Turn on with /rml on in chat
  learning_rate: 0.05     # 0.01 = slow, 0.2 = fast
  max_param_offset: 0.3   # Max drift from base temperature/top_p
  hint_threshold: 3.0     # Negative score before a hint is injected

Preferences persist in ~/.config/mythos/rml_preferences.json across sessions.

---

📜 Command Reference

CLI Commands (terminal)

Command	Description
mythos	Launch chat (default)
mythos chat	Same as above (local, recommended)
mythos cloud	Chat via cloud API (OpenAI-compatible)
mythos cloud set-key <key> --provider nvidia	Save API key with provider (nvidia, openai, together, groq)
mythos cloud status	Show cloud configuration
mythos cloud clear	Remove cloud API key
mythos chat --config <path>	Use a custom config file
mythos web	Launch web UI (Gradio)
mythos web --port 8080 --share	Custom port + public link
mythos init	First-time setup
mythos status	Show config & model status
mythos models	List available GGUF models
mythos config show	Display full resolved configuration
mythos doctor	Diagnose setup issues & dependencies
mythos sessions	List saved session summaries
mythos sessions -n 5	Show last 5 sessions
mythos history	List saved conversations
mythos history -n 10	Show last 10 conversations
mythos scan	Instant static security analysis
mythos scan --deep	AI-powered audit (needs model)
mythos fix --path .	Auto-fix safe patterns (dry-run)
mythos fix --path . --apply	Apply fixes to disk
mythos path add ~/src	Register a codebase
mythos path list	List registered paths
mythos path remove <target>	Remove a registered path
mythos model download	Download default GGUF model
mythos update	Pull latest from GitHub

In-Chat Slash Commands

Command	Description
/help	Show available commands
/config	Show current configuration
/version	Show Mythos version and model info
/tokens	Show token/generation stats
/topp <0.0-1.0>	Set top-p (nucleus sampling)
/topk <int>	Set top-k sampling
/reppen <float>	Set repeat penalty
/maxtokens <int>	Set max tokens per generation
/temp <0.0-2.0>	Set temperature
/persona <name>	Switch persona preset
/compact	Compact conversation context
/copy	Copy last response to clipboard
/rename <name>	Rename current conversation
/export	Export conversation as text
/dump	Dump raw conversation JSON
/wc	Word/char/token count of conversation
/save	Save conversation
/summary	Generate and save a summary
/rml on\|off	Enable/disable RML
/rml good\|bad	Mark last response
/rml stats	Show RML learning stats
/rml reset	Reset RML preferences
/quit	Exit the chat

---

Roadmap

• Multi-model support (swap between GGUF models in-chat)
• MCP (Model Context Protocol) server integration
• LoRA fine-tuning from within the CLI
• RAG over local documents
• Plugin system for community extensions
• Voice input/output via Whisper + Coqui TTS

---

Contributing

We welcome contributions of all kinds — bug fixes, features, docs, and ideas.

1. Fork the repo
2. Create a branch: git checkout -b feature/your-idea
3. Commit your changes: git commit -m "Add your idea"
4. Push: git push origin feature/your-idea
5. Open a Pull Request

See Issues for open tasks. Feel free to open a new issue for bugs, feature requests, or questions.

---

**If OpenMythos-2 speaks to you, leave a star — it helps others find the oracle.** [](https://star-history.com/#creatorofsomethingthatisgood/Open-Mythos-2&Date) [](https://github.com/creatorofsomethingthatisgood/Open-Mythos-2/stargazers) [](https://github.com/creatorofsomethingthatisgood/Open-Mythos-2/network/members)

Disclaimer these answers may not be true by the model.