Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions .github/workflows/desktop.yml
Original file line number Diff line number Diff line change
Expand Up @@ -659,6 +659,11 @@ jobs:
# toolchains (such as Ollama's), which fails the AppImage. Release
# libraries are already stripped.
NO_STRIP: ${{ matrix.platform == 'ubuntu-22.04' && 'true' || '' }}
# The bundled Ollama libraries find each other through the library
# path Ollama sets at runtime, not an rpath, so linuxdeploy can't
# resolve them (e.g. libggml-base.so.0) without this. Only the
# Linux AppImage step reads it.
LD_LIBRARY_PATH: ${{ matrix.platform == 'ubuntu-22.04' && format('{0}/frontend/src-tauri/binaries/ollama-runtime-x86_64-unknown-linux-gnu/build/lib/ollama', github.workspace) || '' }}
with:
projectPath: frontend
tauriScript: npx tauri
Expand Down
16 changes: 10 additions & 6 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,9 @@
[![XERJ Corpus Hub contributor](https://hub.xerj.org/contributors/aryanjsx/badge.svg)](https://hub.xerj.org/contributors/aryanjsx)


[Website](https://aryanjsx.github.io/Locus/) · [Why Locus](#why-locus) · [What it can do](#what-locus-can-do) · [Install](#installation) · [Quick start](#quick-start) · [Privacy](#privacy) · [Docs](https://aryanjsx.github.io/Locus/)
**[Download for Windows, macOS or Linux](https://aryanjsx.github.io/Locus/#download)**

[Website](https://aryanjsx.github.io/Locus/) · [Why Locus](#why-locus) · [What it can do](#what-locus-can-do) · [Install](#installation) · [Quick start](#quick-start) · [Privacy](#privacy) · [Docs](https://aryanjsx.github.io/Locus/getting-started/installation/)

</div>

Expand All @@ -39,14 +41,16 @@ Locus is that stack. It is a framework for local-first personal AI, built around

### Desktop app

Download the installer for your platform from the [**latest release**](https://github.com/aryanjsx/Locus/releases/latest):
The easiest way is the **[download button on the website](https://aryanjsx.github.io/Locus/#download)**. It detects your system and downloads the right installer directly. You can also pick a file from the [latest release](https://github.com/aryanjsx/Locus/releases/latest):

| Platform | File |
|---|---|
| **Windows (64-bit)** | `Locus_<version>_x64-setup.exe` or `Locus_<version>_x64_en-US.msi` |
| **Windows (ARM64)** | `Locus_<version>_arm64-setup.exe` |
| **macOS (Apple Silicon and Intel)** | `Locus_<version>_universal.dmg` |
| **Linux** | `Locus_<version>_amd64.AppImage`, `.deb`, or `.rpm` |
| **Windows 10 and 11 (64-bit)** | `Locus_<version>_x64-setup.exe` or `Locus_<version>_x64_en-US.msi` |
| **Windows 11 on ARM** | `Locus_<version>_arm64-setup.exe` |
| **macOS 10.15 or later (Apple silicon and Intel)** | `Locus_<version>_universal.dmg` |
| **Linux (x86-64)** | `Locus_<version>_amd64.AppImage`, `.deb`, or `.rpm` |

8 GB of memory is enough for a small model; 16 GB runs noticeably better ones. A graphics card helps but isn't required.

On first launch, choose **Ollama** to run models on your PC (or point Locus at your own OpenAI-compatible server). Locus then sets itself up: it installs its private Python runtime, starts the bundled Ollama, and downloads a model sized for your RAM. You don't need to install git, Python, Rust, or Ollama yourself. First launch needs an internet connection and takes a few minutes; later launches start in seconds.

Expand Down
5 changes: 5 additions & 0 deletions docs/assets/favicon.svg
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
223 changes: 6 additions & 217 deletions docs/index.md
Original file line number Diff line number Diff line change
@@ -1,225 +1,14 @@
---
title: Locus
description: Personal AI, On Personal Devices
description: A private AI assistant that runs on your computer. Free and open source for Windows, macOS and Linux.
template: home.html
search:
boost: 2
hide:
- navigation
- toc
---

# Personal AI, On Personal Devices

<p class="hero-tagline">
Locus is a research framework for composable, on-device AI systems.
Build personal AI that runs on your hardware. Cloud APIs are optional.
</p>

<div class="grid cards" markdown>

- :material-image-multiple:{ .lg .middle } **See what people use it for**

---

A gallery of real setups — morning briefs that summarize your overnight Slack and email, a Discord companion that knows your calendar, a code reviewer that works at 30,000 feet. Outcome-first, with links to the docs that explain how to build each one.

[:octicons-arrow-right-24: Browse the Showcase](showcase/index.md)

</div>

---

## Why Locus?

Personal AI agents are exploding in popularity, but nearly all of them still route intelligence through cloud APIs. Your "personal" AI continues to depend on someone else's server. At the same time, [Intelligence Per Watt](https://www.intelligence-per-watt.ai/) research from Stanford's Hazy Research showed that local language models already handle 88.7% of single-turn chat and reasoning queries, with intelligence efficiency improving 5.3× from 2023 to 2025. The models and hardware are increasingly ready. What has been missing is the software stack to make local-first personal AI practical.

Locus is that stack. It is a framework for local-first personal AI, built around three core ideas: shared primitives for building on-device agents; evaluations that treat energy, FLOPs, latency, and dollar cost as first-class constraints alongside accuracy; and a learning loop that improves models using local trace data. The goal is simple: make it possible to build personal AI agents that run locally by default, calling the cloud only when truly necessary. Locus aims to be both a research platform and a production foundation for local AI, in the spirit of PyTorch.

---

## Get Started

=== "Browser App"

Run the full chat UI locally with one script:

```bash
git clone https://github.com/aryanjsx/Locus.git
cd Locus
./scripts/quickstart.sh
```

This installs dependencies, starts Ollama + a local model, launches the backend
and frontend, and opens `http://localhost:5173` in your browser.

=== "Desktop App"

The desktop app is a native window for the Locus UI. It runs everything
on your machine and sets itself up on first launch: its Python runtime,
the bundled Ollama, and a model sized for your RAM.

Download and open the desktop app:

[Download the desktop app](https://github.com/aryanjsx/Locus/releases/latest){ .md-button .md-button--primary }

Available for macOS, Windows and Linux. See the [Downloads](downloads.md) page for which file to pick.

The app connects to `http://localhost:8000` automatically.

!!! warning "macOS first launch"

Run `xattr -cr /Applications/Locus.app` if the app shows as "damaged".

=== "Python SDK"

```python
from locus import Jarvis

j = Jarvis() # auto-detect engine
response = j.ask("Explain quicksort.")
print(response)
```

For more control, use `ask_full()` to get usage stats, model info, and tool results:

```python
result = j.ask_full(
"What is 2 + 2?",
agent="orchestrator",
tools=["calculator"],
)
print(result["content"]) # "4"
print(result["tool_results"]) # [{tool_name: "calculator", ...}]
```

=== "CLI"

```bash
locus ask "What is the capital of France?"

locus ask --agent orchestrator --tools calculator "What is 137 * 42?"

locus serve --port 8000

locus memory index ./docs/
locus memory search "configuration options"
```

---

## Five Primitives for Personal AI

Locus is built around five composable layers. Each has a clean interface and can be swapped independently.

1. **Intelligence** — Pick a model, or let Locus pick one for your hardware. Manages the full catalog of local models across providers.
2. **Engine** — The inference runtime: [Ollama](https://ollama.com), [vLLM](https://github.com/vllm-project/vllm), [SGLang](https://github.com/sgl-project/sglang), [llama.cpp](https://github.com/ggerganov/llama.cpp), cloud APIs, and more. Auto-detects your hardware and recommends the best fit.
3. **Agents** — Multi-step reasoning with tool use. Eight built-in agent types from simple chat to orchestrated workflows.
4. **Tools & Memory** — Web search, calculator, file I/O, code interpreter, retrieval, persistent local state, and any external MCP server.
5. **Learning** — Your AI gets better over time. Every interaction generates traces that drive automatic improvements to model weights, prompts, and agent behavior.

---

## Key Features

<div class="grid cards" markdown>

- **10+ Engine Backends**

---

[Ollama](https://ollama.com), [vLLM](https://github.com/vllm-project/vllm), [SGLang](https://github.com/sgl-project/sglang), [llama.cpp](https://github.com/ggerganov/llama.cpp), [MLX](https://github.com/ml-explore/mlx), [Exo](https://github.com/exo-explore/exo), [LiteLLM](https://github.com/BerriAI/litellm), cloud (OpenAI/Anthropic/Google), and more. Same `InferenceEngine` interface, swap freely.

- **Automated Workflows**

---

Cron-based agents that monitor, summarize, and act. Code review, email triage, research digests — running 24/7 on your hardware.

- **Hardware-Aware**

---

Auto-detects GPU vendor, model, and VRAM. Recommends the optimal engine for your hardware.

- **Offline-First**

---

All core functionality works without a network connection. Cloud APIs are optional extras.

- **OpenAI-Compatible API**

---

`locus serve` starts a FastAPI server with SSE streaming. Drop-in replacement for OpenAI clients.

- **Energy & Cost Tracking**

---

Built-in telemetry for GPU power draw, token costs, and latency. See exactly what each query costs in watts and dollars.

</div>

---

## Documentation

<div class="grid cards" markdown>

- **[Getting Started](getting-started/installation.md)**

---

Install Locus, configure your first engine, and run your first query.

- **[User Guide](user-guide/cli.md)**

---

CLI, Python SDK, and guides for [Morning Digest](user-guide/morning-digest.md), [Deep Research](user-guide/deep-research.md), [Code Assistant](user-guide/code-assistant.md), [Scheduled Monitor](user-guide/scheduled-monitor.md), [Simple Chat](user-guide/chat-simple.md), [Evaluations](user-guide/evaluations.md), agents, memory, tools, and telemetry.

- **[Architecture](architecture/overview.md)**

---

Five-primitive design, registry pattern, query flow, and cross-cutting learning.

- **[API Reference](api-reference/locus/index.md)**

---

Auto-generated reference for every module.

- **[Deployment](deployment/docker.md)**

---

Docker, systemd, launchd. GPU-accelerated container images.

- **[Development](development/contributing.md)**

---

Contributing guide, extension patterns, roadmap, and changelog.

</div>

## Acknowledgments

Locus is a modified version of [OpenJarvis](https://github.com/open-jarvis/OpenJarvis), an open-source framework for local-first personal AI developed at [Hazy Research](https://hazyresearch.stanford.edu/) and the [Scaling Intelligence Lab](https://scalingintelligence.stanford.edu/) at [Stanford SAIL](https://ai.stanford.edu/) as part of [Intelligence Per Watt](https://www.intelligence-per-watt.ai/). OpenJarvis is licensed under Apache 2.0; see [NOTICE](https://github.com/aryanjsx/Locus/blob/main/NOTICE) for details. Locus is an independent project and is not affiliated with or endorsed by Stanford or the OpenJarvis authors.

If you use the underlying research, please cite the OpenJarvis paper:

```bibtex
@misc{saadfalcon2026openjarvispersonalaipersonal,
title={OpenJarvis: Personal AI, On Personal Devices},
author={Jon Saad-Falcon and Avanika Narayan and Robby Manihani and Tanvir Bhathal and Herumb Shandilya and Hakki Orhun Akengin and Gabriel Bo and Andrew Park and Matthew Hart and Caia Costello and Chuan Li and Christopher Ré and Azalia Mirhoseini},
year={2026},
eprint={2605.17172},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2605.17172},
}
```

Read the [OpenJarvis project site](https://openjarvis.stanford.edu/) for the original research motivation, architecture details, and experimental results.
Locus is a free, open-source AI assistant that runs language models on your
own computer, for Windows, macOS and Linux. Download the desktop app, choose
Ollama on first launch, and Locus sets up everything else itself.
Loading
Loading