🤖 Local AI · Built on OpenClaw + Ollama
Jarvis

Your Local AI
Troubleshooting Bot

Jarvis is a private, on-device AI chat assistant that runs entirely on your machine — no internet required, no API fees, no data leaving your computer.


Overview

What Is Jarvis?

Jarvis is a local AI chatbot pipeline — a private troubleshooting assistant built on two open-source tools that handle all routing and inference on your own hardware.

🔒

100% Private

Nothing leaves your machine. No cloud calls, no telemetry, no account required. Your conversations stay local.

Local-firstNo API fees

Fast Responses

Ollama runs quantized models directly on your GPU or CPU — typically 3–20 tokens/sec depending on hardware and model size.

OllamaGPU / CPU
🔧

Model Agnostic

Swap the underlying LLM any time. Mistral, LLaMA 3, Gemma, Phi — any model Ollama supports works out of the box.

MistralLLaMA 3Gemma

Architecture

How It Works

Every message travels through a five-stage pipeline — from your keyboard to the local model and back. Here is the exact sequence of operations:

⌨️
Step 1
User Types
You enter a message or question in the chat interface
🔀
Step 2
OpenClaw Receives
OpenClaw picks up the message and prepares the request payload
🧠
Step 3
Routes to Ollama
OpenClaw forwards the prompt to Ollama’s local REST endpoint
Step 4
Model Generates
The local LLM runs inference and streams the response tokens
💬
Step 5
Response Returns
OpenClaw delivers the completed answer back to your chat interface
Under the hood: OpenClaw is the middleware layer — it manages chat context, formats prompts, and translates between the UI and Ollama’s REST API running on localhost:11434. Ollama handles all model inference locally on your hardware.

Tech Stack

Built With

🦀
Core Middleware

OpenClaw

The routing layer. OpenClaw receives chat input, manages conversation state, and forwards requests to the configured local model endpoint. Lightweight and extensible.

MiddlewareChat RouterLocal
🦙
Inference Engine

Ollama

Runs quantized LLMs locally on your machine via a simple REST API. Supports GPU acceleration when available, falls back gracefully to CPU.

LLM RuntimeREST APIGPU/CPU

Getting Started

How to Run Jarvis

Two steps. Start Ollama to spin up your local model server, then run openclaw from the project directory.

Step 1

Start Ollama

Ollama must be running before you launch Jarvis. It starts a local inference server on port 11434.

# Pull a model if you haven’t yet (one-time setup)
ollama pull mistral

# Start the Ollama server
ollama serve
Ollama runs in the background on localhost:11434. Leave this terminal open.
Step 2

Run OpenClaw

From the Jarvis project directory, launch OpenClaw. It connects to Ollama and opens the chat interface.

# Navigate to the project directory
cd jarvis/

# Launch Jarvis
openclaw
Jarvis is now live in your terminal (or browser if OpenClaw opens a UI). Start chatting — everything runs locally.

You’re Running Jarvis

Once both services are up, Jarvis is live. Ask it to debug errors, explain code, or troubleshoot your system. All responses stay on your machine.

No internet requiredAll localCtrl+C to stop

Status

Project Status

In Progress

Jarvis — Local AI Chat

Core pipeline is functional. Currently refining conversation memory, adding system prompt customization, and improving multi-turn context handling.

OpenClawOllamaLocal LLMPythonFunctional